Beruflich Dokumente
Kultur Dokumente
1)
Administrator Guide
Informatica MDM Multidomain Hub - Version 9.0.1 - 2010
Copyright (c) 2010 Informatica. All rights reserved.
This software and documentation contain proprietary information of Informatica Corporation and are provided under a license
agreement containing restrictions on use and disclosure and are also protected by copyright law. Reverse engineering of the software is
prohibited.
No part of this document may be reproduced or transmitted in any form, by any means (electronic, photocopying, recording or
otherwise) without prior consent of Informatica Corporation. This Software is be protected by U.S. and/or international Patents and
other Patents Pending.
Use, duplication, or disclosure of the Software by the U.S. Government is subject to the restrictions set forth in the applicable software
license agreement and as provided in DFARS 227.7202-1(a) and 227.7702-3(a) (1995), DFARS 252.227-7013©(1)(ii) (OCT 1988), FAR
12.212(a) (1995), FAR 52.227-19, or FAR 52.227-14 (ALT III), as applicable.
The information in this product or documentation is subject to change without notice. If you find any problems in this product or
documentation, please report them to us in writing.
Informatica, Informatica Platform, Informatica Data Services, PowerCenter, PowerCenterRT, PowerCenter Connect, PowerCenter
Data Analyzer, PowerExchange, PowerMart, Metadata Manager, Informatica Data Quality, Informatica Data Explorer, Informatica
B2B Data Transformation, Informatica B2B Data Exchange, Informatica On Demand and Siperian are trademarks or registered
trademarks of Informatica Corporation in the United States and in jurisdictions throughout the world. All other company and product
names may be trade names or trademarks of their respective owners.
Portions of this software and/or documentation are subject to copyright held by third parties, including without limitation: Copyright
DataDirect Technologies. All rights reserved. Copyright © Sun Microsystems. All rights reserved.
This product includes software developed by the Apache Software Foundation (http://www.apache.org/), and other software which is
licensed under the Apache License, Version 2.0 (the "License"). You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0. Unless required by applicable law or agreed to in writing, software distributed under the
License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and limitations under the License.
This product includes software which is licensed under the GNU Lesser General Public License Agreement, which may be found at
http://www.gnu.org/licenses/lgpl.html. The materials are provided free of charge by Informatica, "as-is", without warranty of any kind,
either express or implied, including but not limited to the implied warranties of merchantability and fitness for a particular purpose.
This product includes software which is licensed under the CDDL (the "License"). You may obtain a copy of the License at
http://www.sun.com/cddl/cddl.html. The materials are provided free of charge by Informatica, "as-is", without warranty of any kind,
either express or implied, including but not limited to the implied warranties of merchantability and fitness for a particular purpose. See
the License for the specific language governing permissions and limitations under the License.
This product includes software which is licensed under the BSD License (the "License"). You may obtain a copy of the License at
http://www.opensource.org/licenses/bsd-license.php. The materials are provided free of charge by Informatica, "as-is", without
warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability and fitness for a
particular purpose. See the License for the specific language governing permissions and limitations under the License.
This product includes software Copyright (c) 2003-2008, Terence Parr, all rights reserved which is licensed under the BSD License (the
"License"). You may obtain a copy of the License at http://www.antlr.org/license.html. The materials are provided free of charge by
Informatica, "as-is", without warranty of any kind, either express or implied, including but not limited to the implied warranties of
merchantability and fitness for a particular purpose. See the License for the specific language governing permissions and limitations
under the License.
This product includes software Copyright (c) 2000 - 2009 The Legion Of The Bouncy Castle (http://www.bouncycastle.org) which is
licensed under a form of the MIT License (the "License"). You may obtain a copy of the License at
http://www.bouncycastle.org/licence.html. The materials are provided free of charge by Informatica, "as-is", without warranty of any
kind, either express or implied, including but not limited to the implied warranties of merchantability and fitness for a particular purpose.
See the License for the specific language governing permissions and limitations under the License.
DISCLAIMER: Informatica Corporation provides this documentation "as is" without warranty of any kind, either express or implied,
including, but not limited to, the implied warranties of non-infringement, merchantability, or use for a particular purpose. Informatica
Corporation does not warrant that this software or documentation is error free. The information provided in this software or
documentation may include technical inaccuracies or typographical errors. The information in this software and documentation is
subject to change at any time without notice.
NOTICES
This Informatica product (the “Software”) may include certain drivers (the “DataDirect Drivers”) from DataDirect Technologies, an
operating company of Progress Software Corporation (“DataDirect”) which are subject to the following terms and conditions:
1. THE DATADIRECT DRIVERS ARE PROVIDED “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR
IMPLIED, INCLUDING BUT NOTLIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
PARTICULAR PURPOSE AND NON-INFRINGEMENT.
2. IN NO EVENT WILL DATADIRECT OR ITS THIRD PARTY SUPPLIERS BE LIABLE TO THE END-USER CUSTOMER FOR ANY
DIRECT, INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL OR OTHER DAMAGES ARISING OUT OF THE USE OF THE
ODBC DRIVERS, WHETHER OR NOT INFORMED OF THE POSSIBILITIES OF DAMAGES IN ADVANCE. THESE LIMITATIONS
APPLY TO ALL CAUSES OF ACTION, INCLUDING, WITHOUT LIMITATION, BREACH OF CONTRACT, BREACH OF
WARRANTY, NEGLIGENCE, STRICT LIABILITY, MISREPRESENTATION AND OTHER TORTS.
Contents
Contents 3
Preface 10
Organization 10
Informatica Resources 16
Part 1: Introduction 18
Version Requirements 52
Configuring Datasources 71
-3-
Chapter 5: Building the Schema 73
Sandboxes 216
-4-
About Informatica MDM Hub Processes 218
-5-
Chapter 14: Configuring the Match Process 363
-6-
Running Batch Jobs Using the Batch Viewer Tool 501
-7-
Viewing Custom Java Cleanse Functions 681
Chapter 22: Auditing Informatica MDM Hub Services and Events 684
-8-
About User Access to Hub Console Tools 737
Glossary 744
Index 786
-9-
Preface
Organization
This guide contains the following chapters:
- 10 -
"Configuring the Describes the data landing process and explains how to
Land Process" configure source systems and landing tables.
on page 264
"Configuring the Describes the data staging process and explains how to
Stage Process" configure staging tables, mappings, and other settings in that
on page 274 affect Stage jobs.
"Configuring Explains how to configure data cleansing rules that are run
Data Cleansing" during Stage jobs.
on page 307
"Configuring the Explains how to use the load process, and how to define trust
Load Process" and validation rules.
on page 343
"Configuring the Explains how to configure your Hub Store to match data.
Match Process"
on page 363
"Configuring the Explains how to configure your Hub Store to consolidate
Consolidate data.
Process" on
page 443
"Configuring the Explains how to configure Informatica MDM Hub to write
Publish Process" changes to a message queue.
on page 449
"Executing Describes how to use Hub Console tools to run Informatica
Informatica MDM Hub processes via batch jobs, and how to use third-
MDM Hub party job management tools to schedule and manage
Processes" on Informatica MDM Hub processes via stored procedures.
page 495
"Using Batch Explains how to use the Informatica MDM Hub batch jobs and
Jobs " on page batch groups.
496
"Writing Custom Explains how to schedule Informatica MDM Hub batch jobs
Scripts to using job execution scripts.
Execute Batch
Jobs " on page
559
"Configuring Describes how to use Hub Console tools to configure
Application Informatica MDM Hub client applications that access
Access" on Informatica MDM Hub using Services Integration Framework
page 610 (SIF) requests.
"Generating Describes how to generate ORS-specific SIF request APIs
ORS-specific using the SIF Manager tool in the Hub Console.
APIs and
Message
Schemas" on
page 611
"Setting Up Explains how to set up security for users who will access
Security" on Informatica MDM Hub resources via the Hub Console or
page 621 third-party applications.
"Viewing Explains how to register custom code using the User Object
Registered Registry tool in the Hub Console.
Custom Code"
on page 678
"Auditing Describes how to set up auditing and debugging in the Hub
Informatica Console.
MDM Hub
Services and
Events" on page
684
"Appendixes" Describes other administration-related topics.
- 11 -
on page 697
"Configuring Describes how to configure different character sets for
International internationalization purposes.
Data Support"
on page 698
"Backing Up and Explains how to back up and restore a Informatica MDM Hub
Restoring implementation.
Informatica
MDM Hub" on
page 706
"Configuring Explains how to configure user exits, which are user-
User Exits" on customized, unencrypted stored procedures that are
page 708 configured to execute at a specific point during batch job
execution.
"Viewing Explains how to view details of your Informatica MDM Hub
Configuration implementation using the Enterprise Manager tool in the Hub
Details" on page Console.
715
"Implementing Explains how to add custom buttons to tools in the Hub
Custom Buttons Console that allow users to invoke external services on
in Hub Console demand.
Tools" on page
730
"Configuring Describes how to grant or revoke user access to tools in the
Access to Hub Hub Console using the Tool Access tool.
Console Tools"
on page 737
"Row-level Describes row-level locking, which assists API processing on
Locking" on the Hub concurrent with the execution of batch processes.
page 740
"Glossary" on Defines Informatica MDM Hub terminology.
page 744
- 12 -
Learning About Informatica MDM Hub
Here are the Informatica MDM Hub technical manuals and training materials.
What's New in Informatica MDM Hub describes the new features in this
release.
The Informatica MDM Hub Release Notes contain important information about
this Informatica MDM Hub release. Installers should read the Informatica
MDM Hub Release Notes before installing Informatica MDM Hub.
The Informatica MDM Hub Installation Guide explains to installers how to set
up Informatica MDM Hub, the Hub Store, Cleanse Match Servers, and other
components.
The Informatica MDM Hub Cleanse Adapter Guide explains to installers how to
configure Informatica MDM Hub to use the supported adapters and cleanse
engines.
The Informatica MDM Hub Data Steward Guide explains to data stewards how
to use Informatica MDM Hub tools to consolidate and manage their
organization's data. After reading the Informatica MDM Hub Overview, data
stewards should read the Informatica MDM Hub Data Steward Guide.
- 13 -
Informatica MDM Hub Administrator Guide
The Informatica MDM Hub Metadata Manager Guide explains how to use the
Informatica MDM Hub Metadata Manager tool to validate their organization’s
metadata, promote changes between repositories, import objects into
repositories, export repositories, and related tasks.
The Informatica MDM Hub Resource Kit Guide explains how to install and use
the Informatica MDM Hub Resource Kit, which is a set of utilities, examples,
and libraries that assist developers with integrating the Informatica MDM Hub
into their applications and workflows. This document provides a description of
the sample applications that are included with the Resource Kit.
- 14 -
Informatica Global Customer Support
You can contact a Customer Support Center by telephone or through the
WebSupport Service. WebSupport requires a user name and password. You
can request a user name and password at http://my.informatica.com. Use the
following telephone numbers to contact Informatica Global Customer Support:
- 15 -
Informatica Resources
Informatica Customer Portal
As an Informatica customer, you can access the Informatica Customer Portal
site at http://my.informatica.com. The site contains product information, user
group How-To Library, the Informatica Knowledge Base, the Informatica
Multimedia Knowledge Base information, newsletters, access to the
Informatica customer support case management system (ATLAS), the
Informatica, Informatica Documentation Center, and access to the
Informatica user community.
Informatica Documentation
The Informatica Documentation team takes every effort to create accurate,
usable documentation. If you have questions, comments, or ideas about this
documentation, contact the Informatica Documentation team through email at
infa_documentation@informatica.com. We will use your feedback to improve
our documentation. Let us know if we can contact you regarding your
comments.
- 16 -
Informatica Knowledge Base
As an Informatica customer, you can access the Informatica Knowledge Base
at http://my.informatica.com. Use the Knowledge Base to search for
documented solutions to known technical issues about Informatica products.
You can also find answers to frequently asked questions, technical white
papers, and technical tips. If you have questions, comments, or ideas about
the Knowledge Base, contact the Informatica Knowledge Base team through
email at KB_Feedback@informatica.com.
- 17 -
Part 1: Introduction
Part 1: Introduction
Contents
• "Introduction to Informatica MDM Hub Administration" on page 19
• "Getting Started with the Hub Console" on page 29
- 18 -
Chapter 1: Introduction to
Informatica MDM Hub Administration
Note: This document assumes that you have read the Informatica MDM Hub
Overview and have a basic understanding of Informatica MDM Hub
architecture and key concepts.
Chapter Contents
• "About Informatica MDM Hub Administrators" on page 19
• "Phases in Informatica MDM Hub Administration" on page 20
• "Summary of Administration Tasks" on page 21
For an introduction to using the Hub Console, see "Getting Started with the
Hub Console" on page 29.
- 19 -
Phases in Informatica MDM Hub
Administration
Startup Phase
The startup phase involves installing and configuring core Informatica MDM
Hub components: Hub Store, Hub Server, Cleanse Match Server(s), and
cleanse adapters. For instructions on installing the Hub Store, Hub Server, and
Cleanse Match Servers, see the Informatica MDM Hub Installation Guide for
your platform. For instructions on setting up a cleanse adapter, see the
Informatica MDM Hub Cleanse Adapter Guide.
Note: The instructions in this document assume that you have already
completed the startup phase and are ready to begin configuring your
Informatica MDM Hub implementation.
Configuration Phase
After Informatica MDM Hub has been installed and set up, administrators can
begin configuring and testing Informatica MDM Hub functionality—the data
model and other objects in the Hub Store, data management processes,
external application access, and so on. This phase involves a dynamic,
iterative process of building and testing Informatica MDM Hub functionality to
meet the stated requirements of an organization. The bulk of the material in
this document refers to tasks associated with the configuration phase.
After a schema has been sufficiently built and the Informatica MDM Hub has
been properly configured, developers can build external applications to access
Informatica MDM Hub functionality and resources. For instructions on
developing external applications, see the Informatica MDM Hub Services
Integration Framework Guide.
- 20 -
Production Phase
After a Informatica MDM Hub implementation has been sufficiently configured
and tested, administrators deploy the Informatica MDM Hub in a production
environment. In addition to managing ongoing Informatica MDM Hub
operations, this phase can involve performance tuning to optimize the
processing of actual business data.
Setting Up Security
In this document, "Setting Up Security" on page 621 describes the tasks
associated with setting up security in a Informatica MDM Hub implementation.
Setup tasks vary depending on the particular security requirements of your
Informatica MDM Hub implementation, as described in "Security
Implementation Scenarios" on page 625. Additional security tasks are
involved if external applications access your Informatica MDM Hub
implementation using Services Integration Framework (SIF) requests. For
more information, see "About Setting Up Security" on page 621, "Summary of
Security Configuration Tasks" on page 627, and "Configuration Tasks For
Security Scenarios" on page 628.
- 21 -
Task Usage
on page 662
"Managing Security Required if you are using external security providers
Providers" on page 664 to handle any portion of security in your Informatica
MDM Hub implementation.
"Configuring Access to Required to provide non-administrator users with
Hub Console Tools" on access to Hub Console tools.
page 737
- 22 -
Task Usage
"Configuring Required for creating queries used in packages. For more
Queries" on information, see "About Queries" on page 128 and "Configuring
page 127 Packages" on page 151.
Required for queries used by data stewards in the Merge
Manager tool. For more information, see the Informatica MDM
Hub Data Steward Guide.
"Configuring Required to allow external application users to access
Packages" on Informatica MDM Hub functionality using Services Integration
page 151 Framework (SIF) requests. For more information, see the
Informatica MDM Hub Services Integration Framework Guide.
For more information, see "About Packages" on page 152.
Required to allow data stewards to merge and update records
in the Hub Store using the Merge Manager and Data Manager
tools. For more information, see the Informatica MDM Hub
Data Steward Guide.
To configure the land process for a base object, see "Land Process" on page
221, "Configuring the Land Process" on page 264, and the following topics:
High-Level Tasks for Configuring the Land Process
Task Usage
"Configuring Required to define a unique name internal name for each source
Source system (external applications or systems that provide data to
Systems" on Informatica MDM Hub). For more information, see "About
page 264 Source Systems" on page 265.
"Configuring Required to create landing tables, which provide intermediate
Landing storage in the flow of data from source systems into Informatica
Tables" on MDM Hub. For more information, see "About Landing Tables" on
page 269 page 269.
To configure the stage process for a base object, see "Stage Process" on page
224, "Configuring the Stage Process" on page 274, and the following topics:
High-Level Tasks for Configuring the Stage Process
Task Usage
"Configuring Required to create staging tables, which provide temporary,
Staging Tables" intermediate storage in the flow of data from landing tables
on page 275 into base objects via load jobs. For more information, see
"About Staging Tables" on page 275.
- 23 -
Task Usage
"Mapping Required to enable Informatica MDM Hub to move data from
Columns a landing table to a staging table during the stage process,
Between and also to specify cleanse operations on columns of data
Landing and that are moved. For more information, see "About Mapping
Staging Tables" Columns" on page 286.
on page 286
"Configuring Required to set up data cleansing for a base object during the
Data Cleansing" stage process using the Informatica MDM Hub internal
on page 307 cleanse functionality. For more information, see "About Data
Cleansing in Informatica MDM Hub" on page 307 and the
following topics:
• "Configuring Cleanse Match Servers" on page 308 to
deploy Cleanse Match Servers that execute cleanse
operations and the match process for an Operational
Reference Store (ORS). For more information, see
"About the Cleanse Match Server" on page 308.
• "Configuring Cleanse Lists" on page 333 to specify a
logical grouping of cleanse functions that are executed at
run time in a predefined order. For more information,
see "About Cleanse Lists" on page 333.
• "Using Cleanse Functions" on page 314 to build and
execute cleanse functions that cleanse (standardize or
verify) data. For more information, see "About Cleanse
Functions" on page 314.
To configure the load process for a base object, see "Load Process" on page
227, "Configuring the Load Process" on page 343, and the following topics:
High-Level Tasks for Configuring the Load Process
Task Usage
"Configuring Used when multiple source systems contribute data to a column
Trust for in a base object. Required if you want to designate the relative
Source trust level (confidence factor) for each contributing source
Systems" on system. For more information, see "About Trust" on page 344.
page 344
"Configuring Required if you want to use validation rules to downgrade trust
Validation scores for cell data based on configured conditions. For more
Rules" on information, see "About Validation Rules" on page 353.
page 353
To configure the match process for a base object, see "Match Process" on page
245, "Configuring the Match Process" on page 363, and the following topics:
High-Level Tasks for Configuring the Match Process
Task Usage
"Configuring Match Required for each base object that will be involved in
Properties for a mapping. For more information, see "Match Properties"
Base Object" on on page 367.
page 366
- 24 -
Task Usage
"Configuring Match Required for match column rules involving related
Paths for Related records in either separate tables or in the same table. For
Records" on page more information, see "About Match Paths" on page 373.
373
"Configuring Match Required to specify the base object columns to use in
Columns" on page match column rules. For more information, see "About
387 Match Columns" on page 387.
"Configuring Match Required if you want to use match rule sets to execute
Rule Sets" on page different sets of match column rules at different stages in
399 the match process. For more information, see "About
Match Rule Sets" on page 399.
"Configuring Match Required to specify match column rules that determine
Column Rules for whether two records for a base object are similar enough
Match Rule Sets" to consolidate. For more information, see "About Match
on page 407 Column Rules" on page 408.
"Configuring Required to specify the base object columns (primary
Primary Key Match keys) to use in primary key match rules. For more
Rules" on page 434 information, see "About Primary Key Match Rules" on
page 434.
"Investigating the Useful for investigating the distribution of generated
Distribution of match keys upon completion of the match process. For
Match Keys" on more information, see "About Match Keys Distribution" on
page 438 page 438.
"Configuring Match Required for configuring matches involving non-US
Settings for Non- populations and multiple populations.
US Populations" on
page 699
To configure the publish process for a base object, see "Publish Process" on
page 260, "Configuring the Publish Process" on page 449, and the following
topics:
High-Level Tasks for Configuring the Publish Process
Task Usage
"Configuring Required to specify global settings for all message queues
Global involving outbound Informatica MDM Hub messages.
Message
Queue
Settings" on
page 451
"Configuring Required to set up one or more message queue servers that
Message Informatica MDM Hub will use for incoming and outgoing
Queue messages. The message queue server must already be defined
Servers" on in your application server environment according to the
page 452 application server instructions. For more information, see
"About Message Queue Servers" on page 452.
- 25 -
Task Usage
"Configuring Required to set up one or more outbound message queues for a
Outbound message queue server. For more information, see "About
Message Message Queues" on page 454.
Queues" on
page 454
"Configuring Required for configuring message triggers for a base object.
Message Message queue triggers identify which actions within Informatica
Triggers" on MDM Hub are communicated to outside applications via
page 456 messages in message queues. For more information, see "About
Message Triggers" on page 456.
To execute Informatica MDM Hub processes using tools in the Hub Console,
see "About Informatica MDM Hub Batch Jobs" on page 496, "Using Batch Jobs "
on page 496, and the following topics:
High-Level Tasks for Executing Informatica MDM Hub Process in the Hub Console
Task Usage
"Running Batch Required if you want to run individual batch jobs from the
Jobs Using the Hub Console using the Batch Viewer tool. For more
Batch Viewer information, see "Batch Viewer Tool" on page 501.
Tool" on page
501
"Running Batch Required if you want to run batch jobs in a group from the
Jobs Using the Hub Console, allowing you to configure the execution
Batch Group sequence for batch jobs and to execute batch jobs in parallel.
Tool" on page For more information, see "About Batch Groups" on page
512 512.
- 26 -
High-Level Tasks for Executing Informatica MDM Hub Processes Using Job Management
Tools
Task Usage
"Setting Up Job Required for writing job execution scripts for job
Execution Scripts" management tools. For more information, see "About Job
on page 560 Execution Scripts" on page 560 and "About the C_REPOS_
TABLE_OBJECT_V View" on page 560.
"Monitoring Job Required for determining the execution results of job
Results and execution scripts. For more information, see "Error
Statistics" on page Messages and Return Codes" on page 563 and "Job
563 Execution Status" on page 564.
"Executing Batch Required for executing batch jobs in groups via stored
Groups Using procedures using job scheduling software (such as Tivoli,
Stored CA Unicenter, and so on). For more information, see
Procedures" on "About Executing Batch Groups" on page 598.
page 598
"Developing Required for create, registering, and running custom
Custom Stored stored procedures for batch jobs. For more information,
Procedures for see "About Custom Stored Procedures" on page 604.
Batch Jobs" on
page 604
Configuring Hierarchies
If your Informatica MDM Hub implementation uses Hierarchy Manager to
manage hierarchies, you need to configure hierarchies and their related
objects, including entity icons, entity objects and entity types, relationship
base objects (RBOs) and relationship types, Hierarchy Manager profiles, and
Hierarchy Manager packages. For more information, see "Configuring
Hierarchies" on page 169
- 27 -
Task Usage
"Viewing Used for viewing the following types of user objects that
Registered are registered in the selected ORS: user exits, custom
Custom Code" on stored procedures, custom Java cleanse functions, and
page 678 custom button functions.
"Auditing Used for integration auditing to track activities associated
Informatica MDM with the exchange of data between Informatica MDM Hub
Hub Services and and external systems. For more information, see "About
Events" on page Integration Auditing" on page 684.
684
"Backing Up and Used for backing up and restoring a Informatica MDM Hub
Restoring implementation.
Informatica MDM
Hub" on page 706
"Configuring Required only to configure different character sets in a
International Informatica MDM Hub implementation.
Data Support" on
page 698
"Configuring User Required only if user exits are used. For more information,
Exits" on page see "About User Exits" on page 708.
708
"Viewing Used for remotely monitoring a Informatica MDM Hub
Configuration environment, showing configuration settings for the Hub
Details" on page Server, Cleanse Match Servers, Master Database, and
715 Operational Reference Stores.
"Implementing Used only if you want to create custom buttons for Hub
Custom Buttons Console users to provide on-demand, real-time access to
in Hub Console specialized data services. Applies only to the Merge
Tools" on page Manager, Data Manager, and Hierarchy Manager tools.
730
- 28 -
Chapter 2: Getting Started with the
Hub Console
This chapter introduces the Hub Console and provides a high-level overview of
the tools involved in configuring your Informatica MDM Hub implementation.
Chapter Contents
• "About the Hub Console" on page 29
• "Starting the Hub Console" on page 30
• "Navigating the Hub Console" on page 32
• "Informatica MDM Hub Workbenches and Tools" on page 46
Note: The available tools in the Hub Console depend on your Informatica
license agreement. Therefore, your Hub Console tool might differ from the
previous figure.
- 29 -
Starting the Hub Console
To access the Hub Console:
1. Open a browser window and enter the following URL:
http://YourHubHost:port/cmx/
where YourHubHost is your local Informatica MDM Hub host and port is the
port number. Check with your administrator for the correct port number.
Note: You must use an HTTP connection to start the Hub Console. SSL
connections are not supported.
The Informatica MDM Hub launch screen is displayed.
- 30 -
3. Enter your user name and password.
Note: If you do not have any user names set up, contact Informatica
support.
4. Click OK.
After you have logged in with a valid user name and password,
Informatica MDM Hub will prompt you to choose a target database—the
Master Database or an Operational Reference Store(ORS) with which to
work.
- 31 -
Image Meaning
Unknown. ORS has not been validated since it was initially created,
or since the last time it was updated.
ORS has been validated with no issues. No change has been made
to the ORS since the validation process was made.
ORS has been validated with warnings.
ORS has been validated and errors were found.
For more information, see "About the Hub Store" on page 51.
5. Select the Master Database or the ORS to which you want to connect.
6. Click Connect.
Note: You can easily change the target database once inside the Hub
Console, as described in "Changing the Target Database" on page 37.
The Hub Console screen is displayed (in which the Schema Manager is
selected from the Model workbench).
When you select a tool from the Workbenches page or start a process from the
Processes page, the window is typically divided into several panes:
Pane Description
Workbenches Displays one of the following:
/ Processes • List of workbenches and tools to which you have access (as
shown in the previous figure).
• List of the steps in the process that you are running.
Note: The workbenches and tools that you see depends on what
your company has purchased, as well as to what your
administrator has given you access. If you do not see a
particular workbench or tool when you log into the Hub Console,
then your user account has not been assigned permission to
access it.
Navigation Allows you to navigate items (a list of objects) in the current
Tree tool. For example, in the Schema Manager, the middle pane
contains a list of schema objects (base objects, landing tables,
and so on).
Properties Shows details (properties) for the selected item in the
Panel navigation tree, and possibly other panels if available in the
current tool. Some of the properties might be editable.
- 32 -
Toggling Between the Processes and Workbenches
Views
Informatica MDM Hub groups its tools in two different ways:
Pane Description
By Similar tools are grouped together by workbench—a logical
Workbenches collection of related tools.
By Process Tools are grouped into a logical workflow that walks you
through the tools and steps required for completing a task.
You can click the tabs at the left-most side of the Hub Console window to
toggle between the Processes and Workbenches views.
Note: When you log into Informatica MDM Hub, you see only those
workbenches and processes that contain the tools that your Informatica MDM
Hub security administrator has authorized you to use. The screen shots in this
document show the full set of workbenches, processes, and tools available.
Workbenches View
Processes View
Hub Console displays a list of available processes on the Processes tab. Tools
are organized into common sequences or processes.
- 33 -
Starting a Tool in the Workbenches View
To start a Hub Console tool from the Workbenches view:
1. In the Workbenches view, expand the workbench that contains the tool
that you want to start (see "Informatica MDM Hub Workbenches and Tools"
on page 46).
2. If necessary, expand the workbench node to show the tools associated
with that workbench.
3. Click the tool.
If you selected a tool that requires a different database, the Hub Console
prompts you to select it.
Types of Locks
In the Hub Console, the Write Lock menu provides two types of locks:
Type of Description
Lock
exclusive Allows only one user to make changes to the underlying ORS,
lock preventing any other users from changing the ORS while the
exclusive lock is in effect. For more information, see "Acquiring an
Exclusive Lock" on page 36.
write Allows multiple users to making changes to the underlying
lock metadata at the same time. Write locks can be obtained on the
- 34 -
Type of Description
Lock
Master Database or on an ORS. For more information, see
"Acquiring a Write Lock" on page 36.
Note: The data steward tools—Data Manager, Merge Manager, and Hierarchy
Manager—do not require write locks. For more information about these tools,
see the Informatica MDM Hub Data Steward Guide. The Audit Manager does
not require write locks, either.
The Hub Console takes care of refreshing the lock every 60 seconds on the
current connection. The user can manually release a lock according to the
instructions in "Releasing a Lock" on page 36. If a user switches to a different
database while holding a lock, then the lock is automatically released. If the
Hub Console is terminated, then the lock expires after one minute.
When no locks are in effect in the Hub Console, the Hub Server caches
metadata and other configuration settings for performance reasons. As soon
as a Hub Console user acquires a write lock or exclusive lock, caching is
- 35 -
disabled, the cache is emptied, and Informatica MDM Hub retrieves this
information from the database instead. When all locks are released, caching is
enabled again.
Write locks allow multiple users to edit data in the Hub Console at the same
time. However, write locks do not prevent those users from editing the same
data at the time time. In such cases, the most recently-saved changes prevail.
Releasing a Lock
Clearing Locks
You can force the release of any locks—write or exclusive locks—held by other
users. You might want to do this, for example, to obtain an exclusive lock on
- 36 -
the ORS. Because other users are not warned to save changes before their
write locks are released, you should use this only when necessary.
To change the target database in the Hub Console, do one of the following.
1. On the status bar, click the database name.
Hub Console prompts you to choose a target database with which to work.
For a description of the types of databases that you can select, see
"Starting the Hub Console" on page 30.
2. Select the Master Database or the ORS to which you want to connect.
3. Click Connect.
- 37 -
Changing the Password for a User
To change the password for the currently logged-in user in the Hub Console:
1. From the Options menu, choose Change Password.
2. Specify the password that you want to use instead.
3. Click OK.
Each named object is represented as a node in the hierarchy tree. A node that
contains other nodes is called a parent node. A node that belongs to a parent
node is called a child node.
In the following example in the Schema Manager, the Address base object is
the parent node to the associated child nodes (Columns, Cross-References,
and so on).
The display name is the name of an object as it appears in the navigation tree.
You can change the order in which the objects are displayed in the navigation
tree by clicking Sort By in the tree options area and selecting the appropriate
sort option.
- 38 -
Choose from the following sort options:
• Display Name (a-z) sorts the objects in the tree alphabetically according
to display name.
• Display Name (z-a) sorts the objects in the tree in descending
alphabetical order according to display name.
Filtering Items
You can filter the items shown in the navigation tree by clicking the Filter area
at the bottom of the left pane and selecting the appropriate filter option. The
figures in this section are from the Schema Manager, but the sample
principles apply to other Hub Console tools for which filtering is available.
• If you choose Table, you click the down arrow to display a list of tables
from which to select for your filter.
- 39 -
• Click the Define Item Filter button.
• Select the item(s) that you want to include in the filter, and then click
OK.
Note: Use the No Filter (All Items) option to remove the filter.
Certain Hub Console tools show a View or View By area below the navigation
tree.
• In the Schema Manager, you can show or hide the public Informatica MDM
Hub items by clicking the View area below the navigation tree and
choosing the appropriate command.
- 40 -
• In the Mappings tool, you can view items by mapping, staging table, or
landing table.
• In the Packages tool, you can view items by package or by table.
• In the Users and Groups tool, you can display sub groups and sub users. In
the Batch Viewer, you can group jobs by table, date, or procedure type.
When there is no filter, or when the Some Items filter is selected, Hub Console
displays a Find area above the navigation tree so that you can search for
items by name.
For example, in the Schema Manager, you can search for tables and columns.
1. Click anywhere in the Find area to display the Find window.
- 41 -
2. Type the name (or first few letters of the name) that you want to find.
3. Click the F3 - Find button.
The Hub Console highlights the matched item(s). In the following example,
the Schema Manager displays the list of tables and highlights the table
matches the find criteria.
- 42 -
4. Click anywhere in the Find area to hide the Find window.
For example, in the Schema Manager, you can right-click on certain types of
objects in the navigation tree to see a popup menu of the commands available
for the selected object.
- 43 -
Adding, Editing, and Removing Objects Using
Command Buttons
This section describes generally how you use command buttons to add, edit,
and delete objects in the Hub Console.
Command Buttons
Edit Edit a property for the selected item in the Properties panel.
Indicates that the property is editable.
Delete Remove the selected item.
The following figure shows an example of command buttons on the right side
of the properties panel for the Secure Resources tool.
To see a description about what a command button does, hold the mouse over
the button to display a tooltip.
Adding Objects
To add an object:
1. Acquire a write lock.
- 44 -
2. In the Hub Console tool, select the object whose properties you want to
edit.
3. For each property that you want to edit, click the Edit button next to it,
and specify the new value.
Removing Objects
To remove an object:
1. Acquire a write lock.
2. In the Hub Console tool, select the object that you want to remove.
- 45 -
The Hub Console displays the About Informatica MDM Hub dialog.
3. Click Close.
4. Click Close.
- 46 -
Icon Tool Description
Name
Reference Stores and Datasources" on
page 54.
Users Define users and specify which databases
they can access. Manage global and
individual password policies. Note that
Informatica MDM Hub supports external
authentication for users, such as LDAP. For
more information, see "Configuring
Informatica MDM Hub Users" on page 646.
Security Configure security providers, which are
Providers third-party organizations that provide
security services (authentication,
authorization, and user profile services)
for users accessing Informatica MDM Hub.
For more information, see "Managing
Security Providers" on page 664.
Tool Define which Hub Console tools and
Access processes a user can access. By default,
new user accounts do not have access to
any tools until access is explicitly assigned.
For more information, see "Configuring
Access to Hub Console Tools" on page 737.
Message Define inbound and outbound message
Queues queue interfaces to Informatica MDM Hub.
For more information, see "Configuring the
Publish Process" on page 449.
Metadata Validate Operational Reference Store
Manager (ORS) metadata, promote changes
between repositories, import objects into
repositories, and export repositories. For
more information, see the Informatica
MDM Hub Metadata Manager Guide.
Enterprise View configuration details and version
Manager information for the Hub Server, Cleanse
Servers, the Master Database, and
Operational Reference Stores. For more
information, see "Viewing Configuration
Details" on page 715.
- 47 -
Icon Tool Name Description
Packages Define packages (table views). For more information, see
"Configuring Packages" on page 151.
Cleanse Define cleanse functions to perform on your data. For more
Functions information, see "Using Cleanse Functions" on page 314.
Mappings Map cleansing function outputs to target columns in staging
tables. For more information, see "Mapping Columns
Between Landing and Staging Tables" on page 286.
Hierarchies Set up the structures required to view and manipulate data
relationships in Hierarchy Manager. For more information,
see "Configuring Hierarchies" on page 169.
- 48 -
Icon Tool Description
Name
Batch Execute batch jobs to cleanse, load, match or auto-merge data,
Viewer and view job logs. For more information, see "Running Batch
Jobs Using the Batch Viewer Tool" on page 501.
Cleanse View Cleanse Match Server information, including name, port,
Match server type, and whether server is on or offline. For more
Server information, see "About the Cleanse Match Server" on page
308.
Audit Configure auditing and debugging of application requests and
Manager message queue events. For more information, see "Auditing
Informatica MDM Hub Services and Events" on page 684.
SIF Generate ORS-specific Services Integration Framework (SIF)
Manager request APIs. SIF Manager generates and deploys the code to
support SIF request APIs for packages, remote packages,
mappings, and cleanse functions in an ORS. Once generated,
the ORS-Specific APIs are available as a Web service and via
the Informatica Client JAR. For more information, see
"Generating ORS-specific APIs and Message Schemas" on page
611.
User View registered user exits, user stored procedures, custom
Object Java cleanse functions, and custom GUI functions for an ORS.
Registry For more information, see "Viewing Registered Custom Code"
on page 678.
- 49 -
Part 2: Building the Data Model
Contents
• "About the Hub Store" on page 51
• "Configuring Operational Reference Stores and Datasources" on page 54
• "Building the Schema" on page 73
• "Configuring Queries and Packages" on page 127
• "State Management" on page 159
• "Configuring Hierarchies" on page 169
- 50 -
Chapter 3: About the Hub Store
Chapter Contents
• "Databases in the Hub Store" on page 51
• "How Hub Store Databases Are Related" on page 51
• "Creating Hub Store Databases" on page 52
• "Version Requirements" on page 52
Users for Hub Store databases are created globally—within the Master
Database—and then assigned to specific ORSs. The Master Database also
stores site-level information, such as the number of incorrect log-in attempts
allowed before a user account is locked out.
- 51 -
You can access and manage multiple ORSs from one Master Database. The
Master Database stores the connection settings and properties for each ORS.
Note: An ORS can be registered in only one Master Database. Multiple Master
Databases cannot share the same ORS. A single ORS cannot be associated
with multiple Master Databases.
For more information, see the Informatica MDM Hub Installation Guide.
Version Requirements
Different versions of the Informatica MDM Hub cannot operate together in the
same environment. All components of your installation must be the same
version, including the Informatica MDM Hub software and the databases in the
Hub Store.
- 52 -
If you want to have multiple versions of Informatica MDM Hub at your site,
you must install each version in a separate environment. If you try to work
with a different version of a database, you will receive a message telling you
to upgrade the database to the current version.
- 53 -
Chapter 4: Configuring Operational
Reference Stores and Datasources
Chapter Contents
• "Before You Begin" on page 54
• "About the Databases Tool" on page 54
• "Starting the Databases Tool" on page 55
• "Configuring Operational Reference Stores" on page 55
• "Configuring Datasources" on page 71
- 54 -
Starting the Databases Tool
To start the Databases tool:
1. In the Hub Console, connect to your Master Database. For more
information, see "Changing the Target Database" on page 37.
2. Expand the Informatica Configuration workbench and then click
Databases.
The Hub Console displays the Databases tool (in which a registered ORS is
selected).
- 55 -
Registering an ORS
Note: Registering an ORS will fail if you try to register an ORS that does not
contain the Informatica MDM Hub repository objects or Informatica MDM Hub
procedures.
To register an ORS:
1. Start the Databases tool. For more information, see "Starting the
Databases Tool" on page 55.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
- 56 -
Method Description
Service Connect to Oracle via the service name.
SID Connect to Oracle via the Oracle System ID.
For more information about SERVICE and SID names, see your Oracle
documentation.
5. Accept the connection type you want and choose Next.
The Connection Wizard prompts you to specify connection properties based
on your selected connection type. (Fields in bold are required.)
- 57 -
Service Connection Type
SID Connection Type
- 58 -
Connection Type Properties
Property Description
Database Name for this ORS as it will be displayed in the Hub Con-
Display sole.
Name
Machine Prefix given to keys to uniquely identify records from this
Identifier instance of the Hub Store.
Database IP address or name (if supported on your network) of the
hostname server hosting the Oracle database.
SID Oracle System Identifier (SID) that refers to the
instance of the Oracle database running on the server.
Displayed only if the selected connection type is SID.
Service Name of the Oracle SERVICE used to connect to the
Oracle database. Displayed only if the selected Oracle
Connection Type is Service.
Port The TCP port of the Oracle listener running on the Oracle
database server. The Oracle installation default is 1521.
Oracle Name by which the database is known on your network as
defined in the application server’s TNSNAMES.ORA file. For
TNS example:
Name mydatabase.mycompany.com
This value is set when you install Oracle. See your Oracle
documentation to learn more about this name.
Schema Name of the ORS.
Name
User User name for the ORS. By default, this is the user name that
was specified in the script used to create the ORS. This user
name owns all of the ORS database objects in the Hub Store.
If a proxy user has been configured for this ORS, then you can
specify the proxy user instead. For instructions on creating
ORS databases and defining proxy users, see the Informatica
MDM Hub Installation Guide.
Password Password associated with the User Name for the ORS.
• For Oracle, this password is case-insensitive.
• For DB2, this password is case-sensitive.
By default, this is the password associated with the user name
that was specified in the script used to create the ORS.
If a proxy user has been configured for this ORS, then you
specify the password for the proxy user instead. For
instructions on running of the setup_ors.sql script and
defining proxy users, see the Informatica MDM Hub
Installation Guide.
Note: The Schema Name and the User Name are both the name of the
ORS that was specified in the script used to create the ORS. If you need
this information, consult your database administrator.
6. Specify the connection properties and choose Next.
- 59 -
The Connection Wizard displays a summary of selected connection
properties.
Service Connection Type
SID Connection Type
- 60 -
Additional Connection Properties
Property Description
Connection Connect URL. Default is automatically generated by the
URL Connection Wizard. Format:
Service connection type:
jdbc:oracle:thin:@//database_host:port/service_name
SID connection type:
jdbc:oracle:thin:@//database_host:port/sid
For a service connection type (only), you have the option to
customize and subsequently test a different connection URL.
Example:
jdbc:oracle:thin:@//orclhost:1521/mdmorcl.mydomain.com
Create Check (select) to create the datasource on the application server
datasource after registration. For WebLogic users, you will need to specify
after the WebLogic username and password.
registration
7. For a service connection type, if you want to change the default URL, click
the Edit button. The Connection Wizard prompts you to specify a different
URL:
Specify the URL (it can differ from the URL specified when running the
database creation script described in the Informatica MDM Hub Installation
Guide) and then click OK.
8. If you want to create the datasource on the application server after
registration, check (select) the Create datasource after registration
check box.
Informatica MDM Hub uses the datasources provided by the application
server.
Note for WebLogic: If you are using WebLogic, a dialog box prompts you
for your username and password. This process writes only to the Master
Database. The ORS and datasource need not be available at registration
time.
- 61 -
If you do not check this option, then you will need to manually configure
the datasource, as described in "Configuring Datasources" on page 71.
9. Click OK.
10. Test your database connection settings. For more information, see
"Testing ORS Connections" on page 67.
Note: When you register an ORS that has been used elsewhere, and if the
ORS already has Cleanse Match Servers registered and no other servers
get registered, then you need to re-register one of the Cleanse Match
Servers. This updates the data in c_repos_db_release.
- 62 -
Service Connection Type
- 63 -
Property Description
Store.
Oracle TNS name Name by which the database is known on
your network as defined in the application
server’s TNSNAMES.ORA file.
Password By default, this is the password associated
with the user name that was specified when
the ORS was created.
If a proxy user has been configured for this
ORS, then you specify the password for the
proxy user instead. For instructions on
running of the setup_ors.sql script and
defining proxy users, see the Informatica
MDM Hub Installation Guide.
Update datasource Update the datasource on the appli-
after registration cation server with the updated settings.
6. To update the datasource on the application server with the modified
settings, select (check) the Update datasource after registration
check box.
Note: Updating the datasource settings might cause the JDBC connection
pool settings to be reset to the default values. Be sure to check the JDBC
connection pool settings before and after you click OK so that you can
reapply any customizations to the JDBC connection pool settings.
7. Click OK.
The Databases tool saves your changes.
8. Test your updated database connection settings. For more information,
see "Testing ORS Connections" on page 67.
- 64 -
The following table describes these properties.
Property Description
Database Type Oracle or DB2.
Database ID Identification for the ORS. This ID is used in SIF requests.
The database ID lookup is case-sensitive.
SID Connection Type:
hostname-sid-databasename
Service Connection Type:
servicename-databasename
When registering a new ORS, the host, server, and
database names are normalized.
• Host name is converted to lowercase.
• Database name is converted to uppercase (the
standard for schemas, tables, etc.).
The normalization of each field can be done on a database-
specific basis so that it can be changed if needed.
JNDI Displays the datasource JNDI name for the selected ORS.
Datasource This is the JNDI name that is configured for this JDBC
Name connection on the application server.
SID Connection Type:
jdbc/siperian-hostname-sid-databasename-ds
Service Connection Type:
jdbc/siperian-servicename-databasename-ds
Machine Prefix given to keys to uniquely identify records from this
Identifier instance of the Hub Store.
- 65 -
Property Description
GETLIST Limit Limits the number of records returned through SIF search
(records) requests, such as searchQuery, searchMatch,
getLookupValues, and so on.
Production Specifies whether this ORS is in production mode.
Mode • If not enabled (unchecked, the default), production
mode is disabled, allowing authorized users to edit
metadata for this ORS in the Hub Console.
• If enabled (checked), then production mode is
enabled. Users cannot make changes to the metadata
for this ORS. If a user attempts to acquire a write lock
on an ORS in production mode, the Hub Console will
display a message explaining that the lock cannot be
obtained.
Note: Only Informatica MDM Hub administrator users can
change this setting.
For more information, see "Changing an ORS to Production
Mode" on page 69.
Transition Specifies whether this ORS is running in transition mode.
Available only if Production Mode is enabled for this ORS.
Mode
• If selected (checked), then transition mode is enabled,
allowing users to execute Metadata Manager Promote
actions.
• If not selected (the default), then transition mode is
not enabled.
For more information, see the Informatica MDM Zero
Downtime (ZDT) Install Guide and Informatica MDM Zero
Downtime (ZDT) User Guide.
Batch API Specifies whether this ORS will allow row-level locking for
Interoperability concurrently-executing, asynchronous SIF and batch
operations.
• If selected (checked), then row-level locking is
allowed for asynchronous SIF and batch operations.
• If not selected (the default), row-level locking is
unavailable.
For more information, see "Row-level Locking" on page
740
ZDT Enabled Specifies whether this ORS is running in Zero Downtime
(ZDT) mode.
• If selected (checked), then ZDT is enabled.
• If not selected (the default), then ZDT is not enabled.
For more information, see the Informatica MDM Zero
Downtime (ZDT) Install Guide and Informatica MDM Zero
Downtime (ZDT) User Guide.
4. To change a property, click the button next to it, and edit the
property.
- 66 -
Testing ORS Connections
To test a Hub Store connection to an ORS:
1. Start the Databases tool. For more information, see "Starting the
Databases Tool" on page 55.
2. Select the ORS that you want to test.
Changing Passwords
To change passwords for the Master Database or an ORS, you need to make
changes first on your database server and possibly on your application server
as well.
- 67 -
Changing the Password for an ORS
Encrypting Passwords
To encrypt the new password, execute the following command from the
prompt:
Usage:
java -classpath siperian-common.jar com.siperian.common.security.Blowfish
[key_type] plain_text_password
For example, if admin is your new password, then the command would be:
- 68 -
java -classpath siperian-common.jar com.siperian.common.security.Blowfish
PASSWORD_KEY admin
Plaintext Password: admin
Encrypted Password: A75FCFBCB375F229
Execute the following commands to update the passwords for your ORS and
Master Database:
- 69 -
2. Start the Databases tool. For more information, see "Starting the
Databases Tool" on page 55.
3. Clear any exclusive locks on the ORS.
Note: This setting cannot be changed if the ORS is locked exclusively.
4. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
5. Select the ORS that you want to configure.
The Databases tool displays the database properties for the selected ORS.
6. Change the setting of the Production Mode check box, as described in
"Editing ORS Properties" on page 64.
Select (check) the check box to enable production mode, or clear
(uncheck) it to disable it.
Unregistering an ORS
Unregistering an ORS removes the connection information to this ORS from
the Master Database and removes the datasource definition from the
application server environment.
To unregister an ORS:
1. Start the Databases tool. For more information, see "Starting the
Databases Tool" on page 55.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Select the ORS that you want to unregister.
- 70 -
Configuring Datasources
This section describes how to configure datasources for an ORS. Every ORS
requires a datasource definition in the application server environment.
About Datasources
In Informatica MDM Hub, a datasource specifies properties for an ORS, such
as the location of the database server, the name of the database, the database
user ID and password, and so on. A Informatica MDM Hub datasource points to
a JDBC resource defined in your application server environment. To learn
more about JDBC datasources, see your application server documentation.
Creating Datasources
You might need to explicitly create a datasource if, for example, you created
an ORS using a different application server, or if you did not check (select) the
Create datasource after registration check box when registering the
ORS.
To create a datasource:
1. Start the Databases tool. For more information, see "Starting the
Databases Tool" on page 55.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Right-click the ORS in the Databases list, and then choose Create
Datasource.
Note: If you are running WebLogic, enter the WebLogic user name and
password when prompted.
The Databases tool creates the datasource and displays a progress
message.
- 71 -
4. Click OK.
Removing Datasources
If you have registered an ORS with a configured datasource, you can use the
Databases tool to manually remove its datasource definition from your
application server. After removing the datasource definition, however, the
ORS will still appear in Hub Console. To completely remove a database from
the Hub Console, you need to unregister it (see "Unregistering an ORS" on
page 70).
To remove a datasource:
1. Start the Databases tool. For more information, see "Starting the
Databases Tool" on page 55.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Right-click an ORS in the Databases list, and then choose Remove
Datasource.
Note: If you are running WebLogic, enter the WebLogic user name and
password when prompted.
The Databases tool removes the datasource and displays a progress
message.
4. Click OK.
- 72 -
Chapter 5: Building the Schema
This chapter explains how to design and build your schema in Informatica
MDM Hub.
Chapter Contents
• "Before You Begin" on page 73
• "About the Schema" on page 73
• "Starting the Schema Manager" on page 81
• "Configuring Base Objects" on page 82
• "Configuring Columns in Tables" on page 102
• "Configuring Foreign-Key Relationships Between Base Objects" on page
113
• "Viewing Your Schema" on page 119
Note: The process of designing the schema for your Informatica MDM Hub
implementation is outside the scope of this document. It is assumed that you
have developed a data model—using industry-standard data modeling
methodologies—that is based on a thorough understanding of your
organization’s requirements and in-depth knowledge of the data you are
working with.
- 73 -
of your system, you can have multiple ORS databases in an installation. For
example, you could have a development ORS, a testing ORS, and a production
ORS. For more information, see "About the Hub Store" on page 51 and
"Configuring Operational Reference Stores and Datasources" on page 54.
Before you begin to implement the schema, you must understand the basic
structure of the underlying Informatica MDM Hub schema and its components.
This section introduces the most important tables in an ORS and how they
work together.
Note: You must use tools in the Hub Console to define and manage the
consolidated schema—you cannot make changes directly to the database. For
example, you must use the Schema Manager to define tables and columns.
For details, see "Requirements for Defining Schema Objects" on page 77.
Configurable Tables
The following types of Informatica MDM Hub tables are used to model
business reference data. You must explicitly create and configure these
tables.
Types of Configurable Tables in an ORS
Type Description
of
Table
base Used to store data for a central business entity (such as
object customer, product, or employee) or a lookup table (such
as country or state). In a base object table (or simply a
base object), you can consolidate data from multiple
source systems and use trust settings to determine the
most reliable value of each base object cell. You can
define one-to-many relationships between base objects.
Base objects must be explicitly created and configured
according to the instructions in "Process Overview for
Defining Base Objects" on page 83.
landing Used to receive batch loads from a source system.
table Landing tables must be explicitly created and configured
according to the instructions in "Configuring Landing
Tables" on page 269.
staging Used to load data into a base object. Mappings are
table defined between landing tables and staging tables to
specify whether and how data is cleansed and
standardized when it is moved from a landing table to a
staging table. Staging tables must be explicitly created
and configured according to the instructions in
"Configuring Staging Tables" on page 275.
- 74 -
Infrastructure Tables
The following types of Informatica MDM Hub infrastructure tables are used to
manage and support the flow of data in the Hub Store. Informatica MDM Hub
automatically creates, configures, and maintains these tables whenever you
configure base objects.
Types of Infrastructure Tables in an ORS
Type of Description
Table
cross- Used for tracking the origin of each record in the base object.
reference Named according to the following pattern:
table C_baseObjectName_XREF
where baseObjectName is the root name of the base object (for
example, C_PARTY_XREF). For this reason, this table is sometimes
referred to as the XREF table. When you create a base object,
Informatica MDM Hub automatically creates a cross-reference
table to store information about data coming from source systems.
For more information, see "Cross-Reference Tables" on page 86.
history Used if history is enabled for a base object (see "Enable History"
table on page 90). Named according to the following pattern:
C_baseObjectName_HIST—base object history table, as described
in "Base Object History Tables" on page 89.
C_baseObjectName_HXRF—cross-reference history table, as
described in "Cross-Reference History Tables" on page 89.
where baseObjectName is the root name of the base object (for
example, C_PARTY_HIST and C_PARTY_HXRF).
Informatica MDM Hub creates and maintains several different
history tables to provide detailed change-tracking options,
including merge and unmerge history, history of the pre-cleansed
data, history of the base object, and the cross-reference history.
match key Contains the match keys that were generated for all base object
table records. Named according to the following pattern:
C_baseObjectName_STRP
where baseObjectName is the root name of the base object (for
example, C_PARTY_STRP). For more information, see "Match Key
Tables" on page 240.
match Contains the pairs of matched records in the base object resulting
table from the execution of the match process on this base object.
Named according to the following pattern:
C_baseObjectName_MTCH
where baseObjectName is the root name of the base object (for
example, C_PARTY_MTCH). For more information, see "Populating
the Match Table with Match Pairs" on page 252
external Uses input (C_baseObjectName_EMI) and output (C_
match baseObjectName_EMO) tables.
table • The EMI contains records to match against the records in the
base object.
• The EMO table contains the output data for External Match jobs.
Each row in the EMO represents a pair of matched records—
one from the EMI table and one from the base object:
For more information, see "External Match Jobs" on page 535 and
"External Match Jobs" on page 572.
- 75 -
Type of Description
Table
temporary Informatica MDM Hub creates various temporary tables as needed
tables while processing data (such as during batch jobs). Once the
temporary tables are no longer needed, they are automatically
and periodically removed by a background process.
Type of Description
Relationship
foreign key One base object (the child) contains a foreign key column, which
relationship contains values that match values in the primary key column of
between another base object (the parent). For more information, see
base objects "Process Overview for Defining Foreign-Key Relationships" on
page 114 and "Configuring Foreign-Key Relationships Between
Base Objects" on page 113.
records Within a base object, records are related to each other
within the hierarchically. Allows you to define many-to-many relationships
same base within the base object. For more information, see "Intra-Table
object Paths" on page 377.
Once these relationships are configured in the Hub Console, you can use these
relationships to configure match column rules by defining match paths
between records. For more information, see "Configuring Match Paths for
Related Records" on page 373.
- 76 -
Requirements for Defining Schema Objects
This section describes requirements for configuring schema objects.
In order to make any changes to the schema, you must have a write lock. For
more information, see "Acquiring a Write Lock" on page 36.
Note: To understand which Hub processes create which tables and how to best
manage these tables, please refer to the “Transient Tables” technical note
found on the SHARE portal.
Informatica MDM Hub creates metadata objects that use prefixes and suffixes
added to the names you use for base objects. In order to avoid confusion and
possible data loss, database object names must not use the following strings
as either names or suffixes.
_BVTB _STRPT _TMIN BVLNK_ TCMN_ TGV_
_BVTC _T _TML0 BVTX_ TCMO_ TGV1_
_BVTV _TBKF _TMMA BVTXC_ TCRN_ TLL
_C _TBVB _TMNX BVTXV_ TCRO_ TMA_
_CL _TBVC _TMP0 CLC_ TCSN_ TMF_
_D _TBVV _TMST CSC_ TCSO_ TMMA_
- 77 -
_DLT _TC0 _TNPMA CTL TCVN_ TMR_
_EMI _TC1 _TPMA EXP_ TCVO_ TPBR_
_EMO _TDEL _TPRL GG TCXN_ TRBX_
_HIST _TEMI _TRAW HMRG TCXO_ TUCA_
_HUID _TEMO _TRLG LNK TDCC_ TUCC_
_HXRF _TEMP _TRLT M TDEL_ TUCF_
_JOBS _TEST _TSD PRL TDUMP_ TUCR_
_L _TGA _TSI T_verify_ TUCT_
_LINK _TGA1 _TSNU TBDL_ TFK_ TUCX_
_LMH _TGB _TSTR TBOX_ TFX_ TUDL_
_LMT _TGB1 _TUID TBXR_ TGA_ TUGR_
_MTBM _TGC _TVXR TCBN_ TGB_ TUHM_
_MTCH _TGC1 _VCT TCBO_ TGB1_ TUID_
_MTFL _TMG0 _XREF TCCN_ TGC_ TUK_
_MTFU _TMG1 BV0_ TCCO_ TGC1_ TUPT_
_MVLE _TMG2 BV1_ TCGN_ TGD_ TUTR_
_OPL _TMG3 BV2_ TCGO_ TGF_ TVXRD_
_ORAW _TMGA BV3_ TCHN_ TGM_ TXDL_
_STRP _TMGB BV5_ TCHO_ TGMD_ TXPR_
The following column names are reserved and cannot be used for user-defined
columns.
AFFECTED_LEVEL_CODE ORIG_TGT_ROWID_OBJECT
AFFECTED_ROWID_COLUMN PKEY_SRC_OBJECT
AFFECTED_ROWID_OBJECT PKEY_SRC_OBJECT1
AFFECTED_ROWID_XREF PKEY_SRC_OBJECT2
AFFECTED_SRC_VALUE PREFERRED_KEY_IND
AFFECTED_TGT_VALUE PROMOTE_IND
AUTOLINK_IND PUT_UPDATE_MERGE_IND
AUTOMERGE_IND REPOINTED_IND
CONSOLIDATION_IND ROOT_IND
CREATE_DATE ROU_IND
CREATOR ROWID_GROUP
CTL_ROWID_OBJECT ROWID_JOB
DATA_COUNT ROWID_KEY_CONSTRAINT
DATA_ROW ROWID_MATCH_RULE
- 78 -
DELETED_BY ROWID_OBJECT
DELETED_DATE ROWID_OBJECT_MATCHED
DELETED_IND ROWID_OBJECT_NUM
DEP_PKEY_SRC_OBJECT ROWID_OBJECT1
DEP_ROWID_SYSTEM ROWID_OBJECT2
DIRTY_IND ROWID_SYSTEM
ERROR_DESCRIPTION ROWID_TASK
FILE_NAME ROWID_USER
FIRSTV ROWID_XREF
GENERATED_XREF ROWID_XREF1
GROUP_ID ROWID_XREF2
GVI_NO ROWKEY
HIST_CREATE_DATE RULE_NO
HIST_UPDATE_DATE SDSRCFLG
HSI_ACTION SEQ
HUB_STATE_IND SOURCE_KEY
INTERACTION_ID SOURCE_NAME
INVALID_IND SRC_LUD
LAST_ROWID_SYSTEM SRC_ROWID
LAST_UPDATE_DATE SRC_ROWID_OBJECT
LASTV SRC_ROWID_XREF
LOST_VALUE SSA_DATA
MATCH_REVERSE_IND SSA_KEY
MERGE_DATE STRIP_DATE
MERGE_OPERATION_ID TGT_ROWID_OBJECT
MERGE_UPDATE_NULL_ALLOW_IND TOTAL_BO_IND
MERGE_VIA_UNMERGE_IND TREE_UNMERGE_IND
MRG_SRC_ROWID_OBJECT UNLINK_IND
MRG_TGT_ROWID_OBJECT UNMERGE_DATE
NULL_INDICATOR_BITMAP UNMERGE_IND
NUM_CONTR UNMERGE_OPERATION_ID
OLD_AFFECTED UPDATED_BY
ONLINE_IND WIN_VALUE
ORIG_ROWID_OBJECT_MATCHED XREF_LUD
- 79 -
Other Reserved Words
ADD DECLARE LANGUAGE SELECT
ADMIN DEFAULT LEVEL SEQUENCE
AFTER DELETE LIKE SESSION
ALL DESC MAX SET
ALLOCATE DISTINCT MIN SIZE
ALTER DOUBLE MODIFY SMALLINT
AND DROP MODULE SOME
ANY DUMP NATURAL SPACE
ARRAY EACH NEW SQL
AS ELSE NEXT SQLCODE
ASC END NONE SQLERROR
AT ESCAPE NOT SQLSTATE
AUTHORIZATION EXCEPT NULL START
AVG EXCEPTION NUMERIC STATEMENT
BACKUP EXEC OF STATISTICS
BEFORE EXECUTE OFF SUM
BEGIN EXISTS OLD TABLE
BETWEEN EXIT ON TEMPORARY
BLOB FETCH ONLY TERMINATE
BOOLEAN FILE OPEN THEN
BY FLOAT OPTION TIME
CASCADE FOR OR TO
CASE FOREIGN ORDER TRANSACTION
CHAR FORTRAN OUT TRIGGER
CHARACTER FOUND PLAN TRUNCATE
CHECK FROM PRECISION UNDER
CHECKPOINT FUNCTION PRIMARY UNION
CLOB GO PRIOR UNIQUE
CLOSE GOTO PRIVILEGES UPDATE
COLUMN GRANT PROCEDURE USE
COMMIT GROUP PUBLIC USER
CONNECT HAVING READ USING
CONSTRAINT IF REAL VALUES
CONSTRAINTS IMMEDIATE REFERENCES VARCHAR
CONTINUE IN REFERENCING VIEW
COUNT INDEX RETURN WHEN
- 80 -
CREATE INDICATOR REVOKE WHENEVER
CURRENT INSERT ROLE WHERE
CURSOR INT ROLLBACK WHILE
CYCLE INTEGER ROW WITH
DATABASE INTERSECT ROWS WORK
DATE INTO SAVEPOINT WRITE
DEC IS SCHEMA FALSE
DECIMAL KEY SECTION TRUE
For purely technical reasons, you might want to add columns to a base object.
For example, for a segment match, you must add a segment column. For
more information on adding columns for segment matches, see "Segment
Matching" on page 422.
We recommend that you distinguish columns added to base objects for purely
technical reasons from those added for other business reasons, because you
generally do not want to include these columns in most views used by data
stewards. Prefixing these column names with a specific identifier, such as
CSTM_, is one way to easily filter them out.
For general instructions about using the Schema Manager, see "Navigating the
Hub Console" on page 32. You must use the Schema Manager when defining
- 81 -
tables in an ORS, as described in "Requirements for Defining Schema Objects"
on page 77.
Each individual entity has a single master record—the best version of the
truth—for that entity. An individual entity might have additional records in the
base object (contributing records) that contain the “multiple versions of the
truth” that need to be consolidated into the master record. Consolidation is the
process of merging duplicate records into a single consolidated record that
contains the most reliable cell values from all of the source records.
Important: You must use the Schema Manager to define base objects—you
cannot configure them directly in the database. For more information, see
"Requirements for Defining Schema Objects" on page 77.
- 82 -
Process Overview for Defining Base Objects
To define a base object:
1. Using the Schema Manager, create a base object table according to the
instructions in "Creating Base Objects" on page 95.
The Schema Manager automatically adds system columns, as described in
"Base Object Columns" on page 84.
2. Add the user-defined columns that will contain business data according to
the instructions in "Configuring Columns in Tables" on page 102.
Note: Column names cannot be longer than 26 characters.
3. While configuring column properties, specify which column(s) will use
trust to determine the most reliable value when different source systems
- 83 -
provide different values for the same cell. For more information, see
"Configuring Trust for Source Systems" on page 344.
4. For this base object, create one staging table per source system according
to the instructions in "Configuring Staging Tables" on page 275. For each
staging table, select the base object columns that you want to include.
5. Create any landing tables that you need to store data from source
systems. For more information, see "Configuring Landing Tables" on page
269.
6. Map the landing tables to the staging tables according to the instructions in
"Mapping Columns Between Landing and Staging Tables" on page 286.
If any columns need data cleansing, specify the cleanse function in the
mapping according to the instructions in "Configuring Data Cleansing" on
page 307.
Each staging table must get its data from one landing table (with any
intervening cleanse functions), but the same landing table can provide
data to more than one staging table. Map the primary key column of the
landing table to the PKEY_SRC_OBJECT column in the staging table.
7. Populate each landing table with data using an ETL tool or some other
process, as described in "Land Process" on page 221.
- 84 -
Physical Name Data Description
Type
(Size)
CONSOLIDATION_ INT Integer value indicating the consolidation state of
IND this record. Valid values are:
• 1=unique (represents the best version of the
truth)
• 2=ready for consolidation
• 3=ready for match; this record is a match
candidate for the currently-executing match
process
• 4=available for match; this record is new
(load insert) or has been updated (load
update) and needs to undergo the match
process
• 9=on hold (data steward has put this record on
hold until further notice)
For more information, see "Consolidation Status
for Base Object Records" on page 219.
DELETED_IND INT Reserved for future use.
DELETED_BY VARCHAR Reserved for future use.
(50)
DELETED_DATE DATE Reserved for future use.
LAST_ROWID_ CHAR The identifier of the system responsible for the
SYSTEM (14) most recent update to any cell in the base object
record.
Foreign key referencing ROWID_SYSTEM column
on C_REPOS_SYSTEM table.
DIRTY_IND INT Used to determine whether the tokenize process
generates match keys for this record. Valid values
are:
• 0 = record is up to date
• 1 = record is new or has been updated and
needs to be tokenized
After the record has been tokenized, this flag is
reset to zero (0). For more information, see "Base
Object Records Flagged for Tokenization" on page
243.
INTERACTION_ID INT For state-enabled base objects only. Interaction
identifier that is used to protect a pending cross-
reference record from updates that are not part of
the same process as the original cross-reference
record. For details, see "Protecting Pending
Records Using the Interaction ID" on page 161.
HUB_STATE_IND INT For state-enabled base objects only. Integer value
indicating the state of this record. Valid values
are:
• 0=Pending
• 1=Active (Default)
• -1=Deleted
For details, see "Hub State Indicator" on page
160.
- 85 -
Cross-Reference Tables
This section describes cross-reference tables in the Hub Store.
Each base object has one associated cross-reference table (or XREF table),
which is used for tracking the lineage (origin) of records in the base object.
Informatica MDM Hub automatically creates a cross-reference table when you
create a base object. Informatica MDM Hub uses cross-reference tables to
translate all source system identifiers into the appropriate ROWID_OBJECT
values.
The load process populates cross-reference tables. During load inserts, new
records are added to the cross-reference table. During load updates, changes
are written to the affected cross-reference record(s).
Cross-reference records are visible in the Merge Manager and can be modified
using the Data Manager. For more information, see the Informatica MDM Hub
Data Steward Guide.
- 86 -
Columns in Cross-reference Tables
Cross-reference tables have the following system columns. Note that cross-
reference tables have a unique key representing the combination of the PKEY_
SRC_OBJECT and ROWID_SYSTEM columns.
Physical Name Data Type Description
(Size)
ROWID_XREF NUMBER Primary key that uniquely identifies this record in
(38) the cross-reference table.
PKEY_SRC_ VARCHAR2 Primary key value from the source system. Multi-
OBJECT (255) field/multi-column keys from source systems must
be concatenated into a single key value using the
Informatica MDM Hub internal cleanse process (see
"About Data Cleansing in Informatica MDM Hub" on
page 307) or external cleanse process (an ETL tool
or some other data loading utility).
ROWID_ CHAR (14) Foreign key to C_REPOS_SYSTEM, which is the
SYSTEM Informatica MDM Hub repository table that stores a
Informatica MDM Hub identifier and description of
each source system that can populate the ORS. For
more information, see "Configuring Source
Systems" on page 264.
ROWID_ CHAR (14) Foreign key to the base object. Unique value
OBJECT assigned by Informatica to the associated record in
the base object.
SRC_ LUD DATE Last source update date. Updated only when an
update is received from the source system.
CREATOR VARCHAR2 User or process responsible for creating the cross-
(50) reference record.
CREATE_DATE DATE Date on which the cross-reference record was
- 87 -
Physical Name Data Type Description
(Size)
created.
UPDATED_BY VARCHAR2 User or process responsible for the most recent
(50) update to the cross-reference record.
LAST_ DATE Date of the most recent update to any cell in the
UPDATE_DATE cross-reference record. Can be updated as
applicable during the load and consolidation
processes.
DELETED_IND NUMBER Reserved for future use.
(38)
DELETED_BY VARCHAR2 Reserved for future use.
(50)
DELETED_ DATE Reserved for future use.
DATE
PUT_UPDATE_ NUMBER Indicates whether a record has been edited using
MERGE_IND (38) the Data Manager.
INTERACTION_ NUMBER For state-enabled base objects only. Interaction
ID (38) identifier that is used to protect a pending cross-
reference record from updates that are not part of
the same process as the original cross-reference
record. For more information, see "Protecting
Pending Records Using the Interaction ID" on page
161.
HUB_STATE_ NUMBER For state-enabled base objects only. Integer value
IND (38) indicating the state of this record. Valid values are:
• 0=Pending
• 1=Active (Default)
• -1=Deleted
For more information, see "Hub State Indicator" on
page 160.
PROMOTE_IND NUMBER For state-enabled base objects only. Integer value
(38) indicating the promotion status. Used by the
Promote job to determine whether to promote the
record to an ACTIVE state. Valid values are:
• 0=Do not promote this record
• 1=Promote this record to ACTIVE
This value is not changed to 0 during the Promote
job if the record is not promoted.
For more information, see "Promoting Records
Using the Promote Batch Job" on page 166.
History Tables
This section describes history tables in the Hub Store. If history is enabled for
a base object (see "Enable History" on page 90), then Informatica MDM Hub
maintains history tables for base objects and cross-reference tables. History
tables are used by Informatica MDM Hub to provide detailed change-tracking
options, including merge and unmerge history, history of the pre-cleansed
data, history of the base object, the cross-reference history, and so on.
- 88 -
Base Object History Tables
Item Type
The type of table that you are adding. Select Base Object.
Display Name
The name of this base object as it will be displayed in the Hub Console. Enter a
descriptive name.
Physical Name
The actual name of the table in the database. Informatica MDM Hub will
suggest a physical name for the table based on the display name that you
enter. Make sure that you do not use any reserved name suffixes, as
described in "Rules for Database Object Names" on page 77.
Data Tablespace
The name of the data tablespace. Read-only. For more information, see the
Informatica MDM Hub Installation Guide.
- 89 -
Index Tablespace
The name of the index tablespace. Read-only. For more information, see the
Informatica MDM Hub Installation Guide.
Description
Enable History
When the percentage of the records that have changed is higher than this
value, a complete re-tokenization is performed. If the number of records to be
tokenized does not exceed this threshold, then Informatica MDM Hub deletes
the records requiring re-tokenization from the match key table, calculates the
tokens for those records, and then reinserts them into the match key table.
The default value is 60. For more information, see "Tokenize Process" on page
240.
Note: Deleting can be a slow process. However, if your Cleanse Match Server
is fast and the network connection between Cleanse Match Server and the
database server is also fast, then you may test with a much lower tokenization
threshold (such as 10%). This will enable you to determine whether there are
any gains in performance.
- 90 -
Duplicate Match Threshold
This parameter is used only with the Match for Duplicate Data job for initial
data loads. The default value is 0. To enable this functionality, this value must
be set to 2 or above. For more information, see "Match for Duplicate Data
Jobs" on page 552 and the Informatica MDM Hub Data Steward Guide.
The load process inserts and updates batches records in the base object. The
load batch size specifies the number of records to load per batch cycle (default
is 1000000). For more information, see "Loading Records by Batch" on page
231, and "Configuring the Load Process" on page 343.
This specifies the execution timeout (in minutes) when executing a match
rule. If this time limit is reached, then the match process (whenever a match
rule is executed, either manually or via a batch job) will exit. If a match
process is executed as part of a batch job, the system should move onto the
next match. It will stop if this is a single match process. The default value is
20. Increase this value only if the match rule and data are very complex.
Generally, rules are able to complete with 20 minutes (the default). For more
information, see "Match Process" on page 245 and "Configuring the Match
Process" on page 363.
Parallel Degree
Oracle only. This specifies the degree of parallelism set on the base object
table and its related tables. It does not take effect for all batch processes, but
can have a beneficial effect on performance when it is used. However, its use
is constrained by the number of CPUs on the database server machine, as well
as the amount of memory available. The default value is 1.
If this value is greater than zero, when parents are merged, the related child
records are set as unconsolidated. If set, when parents are merged, then
related child records are flagged as New again (consolidation indicator is 4,
see "Consolidation Status for Base Object Records" on page 219) so that they
can be matched. The default value is 0. For more information, see
"Consolidation Indicator" on page 219 and "Immutable Rowid Object" on page
443.
- 91 -
Generate Match Tokens on Load
Once the child match/merge is complete, you can run the match process on
the parent to force it to tokenize. This is also useful in cases where you have a
limited window in which to perform the load process. Not tokenizing will save
time in the load process, at the cost of tokenizing the data later.
You must tokenize before you match your data. For more information, see
"Load Process" on page 227, "Generating Match Tokens (Optional)" on page
239, and "Generating Match Tokens During Load Jobs" on page 544.
You can PUT data into a base object using the Data Manager (see the
Informatica MDM Hub Data Steward Guide). If you are using the Data
Manager to PUT data, you can enable (check) this value to tokenize your data
later. Performing this operation later allows you to process PUT requests
faster. Use this only when you know that the data will not be matched
immediately. For more information, see "Tokenize Process" on page 240.
Note: Do not use the Generate Match Tokens on Put option if you are using the
SIF API. If you have this parameter enabled, your SIF Put and CleansePut
requests will fail. Use the Tokenize request instead. Enable Generate Match
Tokens on Put only if you are not using the SIF API and you want data steward
updates from the Hub Console to be tokenized immediately. For more
information, see "Editing Base Object Properties" on page 95.
- 92 -
For more information, see "Match Process" on page 245 and "Configuring the
Match Process" on page 363.
Specifies the maximum number of seconds that a SIF request will wait to
obtain a row-level lock. Applies only if row-level locking is enabled for an
ORS, as described in "Enabling Row-level Locking on an ORS" on page 741. For
more information, see "Row-level Locking" on page 740.
Specifies the maximum number of seconds that a batch job will wait to obtain
a row-level lock. Applies only if row-level locking is enabled for an ORS, as
described in "Enabling Row-level Locking on an ORS" on page 741. For more
information, see "Row-level Locking" on page 740.
Specifies whether Informatica MDM Hub manages the system state for
records in this base object. By default, state management is disabled. Select
(check) this check box to enable state management for this base object in
support of approval workflows. If enabled, this base object is referred to in
this document as a state-enabled base object. For more information, see
"State Management" on page 159 and "Enabling State Management" on page
163.
Note: If the base object has custom query, when you disable state
management on the base object, you will always get a warning pop-up
window, even when the hub_state_ind is not included in the custom query.
- 93 -
• A link-style base object is used with Informatica MDM Hub’s match and
link capabilities. If selected, Informatica MDM Hub creates a LINK table for
this base object.
If you change a link-style base object back to a merge-style base object,
the Schema Manager prompts you to confirm whether you want to drop the
LINK table.
Lookup Indicator
Specifies how values are retrieved in the Informatica MDM Hub Business Data
Director.
• If selected (enabled), then the Business Data Director displays drop-down
lists of lookup values.
• If not selected (disabled), then the Business Data Director displays a
search wizard that prompts users to select a value from a data table.
- 94 -
Creating Base Objects
To create each base object in your schema:
1. Start the Schema Manager according to the instructions in "Starting the
Schema Manager" on page 81.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Right-click in the left pane of the Schema Manager and choose Add Item
from the popup menu.
The Schema Manager displays the Add Table dialog box.
4. Specify the basic base object properties. For more information, see "Basic
Base Object Properties" on page 89.
5. Click OK.
The Schema Manager creates the new base table in the Operational
Reference Store (ORS), along with any support tables, and then adds the
new base object table to the schema tree.
- 95 -
4. For each property that you want to edit on the Basic tab, click the Edit
button next to it, and specify the new value. For more information, see
"Basic Base Object Properties" on page 89.
5. If you want, check (select) the Enable History check box to have
Informatica MDM Hub keep a log of records that are inserted, updated, or
deleted. You can use a history table for audit purposes.
6. To modify other base object properties, click the Advanced tab.
- 96 -
7. Specify the advanced properties for this base object. For more
information, see "Advanced Base Object Properties" on page 90.
8. In the left pane, click Match/Merge Setup beneath the base object’s name.
When you configure columns for a base object, system indexes are created
automatically for primary keys and unique columns. In addition, Informatica
MDM Hub automatically drops and creates system indexes as needed when
executing batch jobs or stored procedures.
- 97 -
A custom index is a optional, supplemental index for a base object that you
can define and have Informatica MDM Hub maintain automatically. Custom
indexes are non-unique.
You might want to add a custom index to a base object for performance
reasons. For example, suppose an external application calls the SIF
SearchQuery request to search a base object by last name. If the base object
has a custom index on the last name column, the last name search is
processed more quickly. For custom indexes that are registered in
Informatica MDM Hub, custom indexes are automatically dropped and
recreated during batch execution to improve performance.
You have the option to manually define indexes outside the Hub Console using
a database utility for your database platform. For example, you could create a
function-based index—such as Upper(Last_Name) in the index expression—in
support of some specialized operation. However, if you add a user-defined
index which are not supported by the Schema Manager, then the custom index
is not registered with Informatica MDM Hub, and you are responsible for
maintaining that index—Informatica MDM Hub will not maintain it for you. If
you do not properly maintain the index, you risk affecting batch processing
performance.
- 98 -
1. In the Schema Manager, navigate to the Custom Index Setup node for the
base object that you want to work with, as described in "Navigating to the
Custom Index Setup Node" on page 98.
If an index already exists for the selected column(s), the Schema Manager
displays an error message and does not create the index.
- 99 -
Click OK to close the dialog box.
To change a custom index, you must delete the existing custom index and add
a new custom index with the columns that you want.
- 100 -
5. Click Close.
- 101 -
Configuring Columns in Tables
After you have created a table (base object or landing table), you use the
Schema Manager to define the columns for that table according to the
"Requirements for Defining Schema Objects" on page 77. You must use the
Schema Manager to define columns in tables—you cannot configure them
directly in the database.
Note: In the Schema Manager, you can also view the columns for cross-
reference tables and history tables, but you cannot edit them.
About Columns
This section provides general information about table columns.
For more information about system columns in Hub Store tables, see:
• "Base Object Columns" on page 84
• "Columns in Cross-reference Tables" on page 87
• "History Tables" on page 88
• "Building the Schema" on page 73
• "Landing Table Columns" on page 269
• "Staging Table Columns" on page 275
Informatica MDM Hub uses a common set of data types for columns that map
directly to the following Oracle and DB2 data types.
- 102 -
Note: For information regarding the available data types, refer to the product
documentation for your database platform.
Informatica MDM Hub Data Type Oracle Data Type DB2 Data Type
CHAR CHAR CHAR
VARCHAR VARCHAR2 VARCHAR
NVARCHAR2 NVARCHAR2
NCHAR NCHAR
DATE DATE DATE
NUMBER NUMBER NUMERIC
INT INTEGER INT or INTEGER
Column Properties
- 103 -
Property Description
insert into this base object fails. To use this feature, you must have
unique keys across all systems.
Validate Enable (check) this option if validation rule(s) will be configured for
this column. Validation rules are applied during the load process to
downgrade trust scores for cell values in this column. For more
information, see "Enabling Validation Rules for a Column" on page
354.
Apply Determines the survivorship of null values for put operations and
Null during the consolidation process.
Values • By default, this option is disabled. Trust scores for cells
containing null values are automatically downgraded so that,
during put operations or consolidation, null values are unlikely to
win over non-null values. Instead, non-null values from the next
available trusted source would survive.
- 104 -
identifiers (social security number, tax ID number, driver’s license
number, and so on).
In the Schema Manager, you can define multiple GBID columns in a base
object. For example, an employee table might have columns for social
security number and driver’s license number, or a vendor table might have a
tax ID number.
GBIDs help with the traceability of your data. Traceability is keeping track of
the data so that you can determine its lineage—which systems, and which
records from those systems, contributed to consolidated records. When you
define GBID columns in a base object, the Schema Manager creates a
separate table for this base object (the table name ends with _HUID) that
tracks the old and new values (current/obsolete value pairs).
For example, suppose two of your customers (both of which had different tax
ID numbers) merged into a single company, and one tax ID number survived
while the other one became obsolete. If you defined the taxID number column
as a GBID, Informatica MDM Hub could help you track both the current and
historical tax ID numbers so that you could access data (via SIF requests)
using the historical value.
Note: Informatica MDM Hub does not perform any data verification or error
detection on GBID columns. If the source system has duplicate GBID values,
then those duplicate values will be passed into Informatica MDM Hub.
- 105 -
Columns in Staging Tables
The columns for staging tables cannot be defined using the column editor.
Staging table columns are a special case, as they are based on some or all
columns in the staging table’s target object. You use the Add/Edit Staging
Table window to select the columns on the target table that can be populated
by the staging table. Informatica MDM Hub then creates each staging table
column with the same data types as the corresponding column in the target
table. See "Configuring Staging Tables" on page 275 for more information on
choosing the columns for staging tables.
A base object cannot have more than 200 user-defined columns if it will have
match rules that are configured for automatic consolidation. For more
information, see "Flagging Matched Records for Automatic or Manual
Consolidation" on page 254 and "Specifying Consolidation Options for Matched
Records" on page 408.
Note: In the above example, the schema shows ANSI SQL data types that
Oracle converts to its own data types. For more information, see "Data
Types for Columns" on page 102.
- 106 -
Command Buttons in the Column Editor
The Properties pane in the Column Editor contains the following command
buttons:
Button Name Description
Add Add new columns. For more information, see "Adding
Columns" on page 108.
Delete Remove existing columns. For more information, see
"Deleting Columns" on page 112.
Move Move the selected column up in the display order. For more
Up information, see "Changing the Column Display Order" on
page 112.
Move Move the selected column down in the display order. For more
Down information, see "Changing the Column Display Order" on
page 112.
Import Add new columns by importing column definitions from
another table. For more information, see "Importing Column
Definitions From Another Table" on page 109.
Expand Expand the table columns view. For more information, see
View "Expanding the Table Columns View" on page 107.
Restore Restore the table columns view. For more information, see
View "Expanding the Table Columns View" on page 107.
Save Saves changes to the column definitions.
You can toggle the Show System Columns check box to show or hide system
columns. For more information, see "Types of Columns in ORS Tables" on
page 102.
You can expand the properties pane to display all the column properties in a
single pane. By default, the Schema Manager displays column definitions in a
contracted view.
- 107 -
• Click the button.
Adding Columns
To add a column:
1. Navigate to the column editor for the table that you want to configure. For
more information, see "Navigating to the Column Editor" on page 106.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
4. For each column, specify its properties. For more information, see
"Column Properties" on page 103.
- 108 -
Importing Column Definitions From Another Table
To import some of the column definitions from another table:
1. Navigate to the column editor for the table that you want to configure. For
more information, see "Navigating to the Column Editor" on page 106.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
4. Specify the connection properties for the schema that you want to import.
If you need more information about the connection information to specify
here, contact your database administrator.
The settings for the User name / Password fields depend on whether proxy
users are configured for your Informatica MDM Hub implementation.
• If proxy users are not configured (the default), then the user name will
be the same as the schema name.
• If proxy users are configured, then you must specify the custom user
name / password so that Informatica MDM Hub can use those
credentials to access the schema.
For more information about proxy user support, see the Informatica MDM
Hub Installation Guide.
5. Click Next.
Note: The database you enter does not need to be the same as the
Informatica ORS that you’re currently working in, nor does it need to be a
Informatica ORS.
- 109 -
The only restriction is that you cannot import from a relational database
that is a different type from the one in which you are currently working.
For example, if your database is an Oracle database, then you can import
columns only from another Oracle database.
The Schema Manager displays a list of the tables that are available for
import.
10. Click the Save button to save the column(s) that you have added.
- 110 -
• reduce the length of a CHAR, VARCHAR, NCHAR, or NVARCHAR2 field
• change the scale or precision of a NUMBER field
Important: As with any schema changes that are attempted after the tables
have been populated with data, manage changes to columns in a planned and
controlled fashion, and ensure that the appropriate database backups are done
before making changes.
- 111 -
Changing the Column Display Order
You can move columns up or down in the display order. Changing the display
order does not affect the physical table in the database.
• Click the button to move the selected column down in the display
order.
Deleting Columns
Removing columns should be approached with extreme caution. Any data that
has already been loaded into a column will be lost when the column is
removed. It can also be a slow process due to the number of underlying tables
that could be affected. You must save the changes immediately after
removing the existing columns.
- 112 -
Configuring Foreign-Key Relationships
Between Base Objects
This section describes how to configure foreign key relationships between
base objects in your Informatica MDM Hub implementation. For a general
overview of foreign key relationships, see "Process Overview for Defining
Foreign-Key Relationships" on page 114. For more information about parent-
child relationships, see "Configuring Match Paths for Related Records" on page
373.
- 113 -
Process Overview for Defining Foreign-Key
Relationships
To create a foreign-key relationship:
1. Create the parent table. For more information, see "Creating Base
Objects" on page 95.
2. Create the child table. For more information, see "Deleting Base Objects"
on page 101.
3. Define the foreign key relationship between them according to the
instructions in "Adding Foreign-Key Relationships" on page 115.
If the child table contains generated keys from the parent table, the load
process copies the appropriate primary key value from the parent table into
the child table.
- 114 -
Adding Foreign-Key Relationships
To add a foreign-key relationship between two base objects:
1. Start the Schema Manager according to the instructions in "Starting the
Schema Manager" on page 81.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. In the schema tree, expand a base object (the base object that will be the
child in the relationship).
4. Right-click Relationships.
The Schema Manager displays the Properties tab of the Relationships
page.
- 115 -
8. Click OK.
9. Click the Diagram tab to view the foreign-key relationship diagram.
Note: After you have created a relationship, if you go back and try to create
another relationship, the column is not displayed because it is in use. When
you delete the relationship, the column will be displayed.
To edit the lookup display name for a foreign-key relationship between two
base objects:
1. Start the Schema Manager according to the instructions in "Starting the
Schema Manager" on page 81.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. In the schema tree, expand a base object and right-click Relationships.
The Schema Manager displays the Properties tab of the Relationships
page.
- 116 -
4. On the Properties tab, click the foreign-key relationship whose properties
you want to view.
The Schema Manager displays the relationship details.
5. Click the Edit button next to the Lookup Display Name and specify the
new value.
6. If you want, select the Has Associated Index check box to add an index
on this foreign key relationship, or clear it to remove an existing index.
- 117 -
MDM Hub automatically defines and enforces to protect the referential
integrity of your schema.
- 118 -
Viewing Your Schema
You can use the Schema Viewer tool in the Hub Console to visualize the
schema in an ORS. The Schema Viewer is particularly helpful for visualizing a
complex schema.
The Diagram Pane in the Schema Viewer contains the following command
buttons:
Button Name Description
Zoom Zooms in and magnifies a smaller area of the schema
In diagram, as described in "Zooming In" on page 120.
Zoom Zooms out and displays a larger area of the schema diagram,
Out as described in "Zooming Out" on page 121.
- 119 -
Button Name Description
Zoom Zooms out to displays the entire schema diagram, as
All described in "Zooming All" on page 121.
Layout Toggles between a hierarchic and orthogonal view, as
described in "Switching Views of the Schema Diagram" on
page 121.
Options Shows or hides column names and controls the orientation of
the hierarchic view, as described in "Configuring Schema
Viewer Options" on page 123.
Save Saves the schema diagram as a JPG file, as described in
"Saving the Schema Diagram as a JPG Image" on page 124.
Print Prints the schema diagram, as described in "Printing the
Schema Diagram" on page 125.
Zooming In
Note that the gray highlight box in the Overview Pane has grown smaller to
indicate the portion of the schema that is displayed in the diagram pane.
- 120 -
Zooming Out
Note that the gray box in the Overview Pane has grown larger to indicate a
larger viewing area.
Zooming All
To zoom all of the schema diagram, which means that the entire schema
diagram is displayed in the Diagram Pane:
• Click the button.
The Schema Viewer zooms out to display the entire schema diagram.
Hierarchic View
The following figure shows an example of the hierarchic view (the default).
- 121 -
Orthogonal View
The following figure shows the same schema in the orthogonal view.
Toggling Views
- 122 -
The Schema Viewer displays the other view.
- 123 -
Pane Description
Show Controls whether column names appear in the entity boxes.
column • Check (select) this option to display column names in the
names entity boxes.
• Uncheck (clear) this option to hide column names and
display only entity names in the entity boxes.
Orientation Controls the orientation of the schema hierarchy. One of the
following values:
• Top to Bottom (default)—Hierarchy goes from top to
bottom, with the highest-level node at the top.
• Bottom to Top—Hierarchy goes from bottom to top,
with the highest-level node at the bottom.
• Left to Right—Hierarchy goes from left to right, with the
highest-level node at the left.
• Right to Left—Hierarchy goes from right to left, with the
highest-level node at the right.
In the following example, column names are hidden.
3. Click OK.
- 124 -
2. Navigate to the location on the file system where you want to save the JPG
file.
3. Specify a descriptive name for the JPG file.
4. Click Save.
The Schema Viewer saves the file.
Pane Description
Print Scope of what to print:
Area • Print All—Print the entire schema diagram.
• Print viewable—Print only the portion of the schema diagram
that is currently visible in the Diagram Pane.
Page Page output options, such as media, orientation, and margins.
Settings
Printer Printer options based on available printers in your environment.
Settings
3. Click Print.
The Schema Viewer sends the schema diagram to the printer.
- 125 -
- 126 -
Chapter 6: Configuring Queries and
Packages
Chapter Contents
• "Before You Begin" on page 127
• "About Queries and Packages" on page 127
• "Configuring Queries" on page 127
• "Configuring Packages" on page 151
Configuring Queries
This section describes how to create and modify queries using the Queries tool
in the Hub Console. The Queries tool allows you to create simple, advanced,
and custom queries.
- 127 -
About Queries
In Informatica MDM Hub, a query is a request to retrieve data from the Hub
Store. Just like any SQL-based query statement, Informatica MDM Hub
queries allow you to specify, via the Hub Console, the criteria used to retrieve
that data—tables and columns to include, conditions for filtering records, and
sorting and grouping the results. Queries that you save in the Queries tool can
be used in packages, and data stewards can use them in the Data Manager and
Merge Manager tools.
Query Capabilities
Types of Queries
Queries are dependent on the base object columns from which they retrieve
data. If changes are made to the column configuration in the base object
associated with a query, then the queries—including custom queries—are
updated automatically. For example, if a column is renamed, then the name is
updated in any dependent queries. If a column is deleted in the base object,
then the consequences depend on the type of query:
• For a custom query, the query becomes invalid and must be manually
fixed in the Queries tool or the Packages tool. Otherwise, if executed, an
invalid query will return an error.
• For all other queries, the column is removed from the query, as well as
from any packages that depend on the query.
- 128 -
Starting the Queries Tool
To start the Queries tool:
• Expand the Model workbench and then click Queries.
The Hub Console displays the Queries tool.
- 129 -
4. Enter a descriptive name for this query group.
5. Enter a description for this query group.
6. Click OK.
The Queries tool adds the new query group to the tree.
4. For each property that you want to edit, click the Edit button next to it,
and specify the new value.
You can delete an empty query group but not a query group that contains
queries.
Configuring Queries
This section describes how to configure queries.
Adding Queries
To add a query:
- 130 -
1. In the Hub Console, start the Queries tool according to the instructions in
"Starting the Queries Tool" on page 129.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Select the query group to which you want to add the query.
4. Right-click in the Queries pane and choose New Query.
The Queries tool displays the New Query Wizard.
5. If you see a Welcome screen, click Next.
Property Description
Query name Descriptive name for this query.
Description Option description of this query.
Query Group Select the query group to which this query belongs.
Select primary table Primary table from which this query retrieves data.
7. Do one of the following:
• If you want the query to retrieve all columns and all records from the
primary table, click Finish to complete the process of creating the
query.
• If you want to specify selection criteria, click Next and continue.
The Queries tool displays the Select query columns window.
- 131 -
8. Select the query columns from which you want the query to retrieve data.
Note: PUT-enabled packages require the Rowid Object column in the
query.
9. Click Finish.
The Queries tool adds the new query to the tree.
10. Refine the query criteria by proceeding to the instructions in "Editing Query
Properties" on page 132.
Once you have created a query, you can modify its properties to refine the
criteria it uses to retrieve data from the ORS.
- 132 -
The properties pane displays the following set of tabs:
Tab Description
Tables Tables associated with this query. Corresponds to the SQL
FROM clause. For more information, see "Configuring the
Table(s) in a Query" on page 134.
Select Columns associated with this query. Corresponds to the SQL
SELECT clause. For more information, see "Configuring the
Column(s) in a Query" on page 136.
Conditions Conditions associated with this query. Determines selection
criteria for individual records. Corresponds to the SQL WHERE
clause. For more information, see "Configuring Conditions for
Selecting Records of Data" on page 139.
Sort Sort order for the results of this query. Corresponds to the SQL
ORDER BY clause. For more information, see "Specifying the
Sort Order for Query Results" on page 142.
Grouping Grouping for the results of this query. Corresponds to the SQL
GROUP BY clause. "Specifying the Grouping for Query Results"
on page 144.
SQL Displays the SQL associated with the selected query settings.
"Viewing the SQL for a Query" on page 147.
4. Make the changes you want.
- 133 -
Configuring the Table(s) in a Query
The Tables tab displays the table(s) from which the query will retrieve
information. The information in this tab corresponds to the SQL FROM clause.
6. If prompted, select a foreign key relationship (if you want), and then click
OK.
The Queries tool displays the added table in the Tables tab.
- 134 -
For multiple tables, the Queries tool displays all added tables in the Tables
tab.
A query must have multiple tables in order for you to remove a table. You
cannot remove the last table in a query.
- 135 -
Configuring the Column(s) in a Query
The Select tab displays the list of column(s) in one or more source tables from
which the query will retrieve information. The information in this tab
corresponds to the SQL SELECT clause.
5. Expand the list for the table containing the column that you want to add.
- 136 -
The Queries tool displays the list of columns for the selected table.
- 137 -
Changing the Column Order
To change the order in which the columns will appear in the result set (if the
list contains multiple columns):
1. In the Hub Console, start the Queries tool according to the instructions in
"Starting the Queries Tool" on page 129.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click the Select tab.
4. Select one column that you want to move.
5. Do one of the following:
Adding Functions
You can add aggregate functions to your queries (such as COUNT, MIN, or
MAX). At run time, these aggregate functions appear in the usual syntax for
the SQL statement used to execute the query—such as:
select col1, count(col2) as c1 from table_name group by col1
- 138 -
7. Click OK.
Adding Constants
The Conditions tab displays a list of condition(s) that the query will use to
select records from the table. A comparison is a query condition that involves
one column, one operator, and either another column or a constant value. The
information in this tab corresponds to the SQL WHERE clause.
Operators
- 139 -
Operator Description
= Equals.
<> Does not equal.
IS NULL
IS NOT NULL
LIKE Value in the comparison column must be like the search value
(includes column values that match the search value). For example,
if the search value is %JO% for the last_name column, then the
parameter will match column values like “Johnson”, “Vallejo”,
“Major”, and so on.
NOT Value in the comparison column must not be like the search value
LIKE (excludes column values that match the search value). For
example, if the search value is %JO% for the last_name column,
then the parameter will omit column values like “Johnson”,
“Vallejo”, “Major”, and so on.
< Less than.
<= Less than or equal to.
> Greater than.
>= Greater than or equal to.
Adding a Comparison
- 140 -
8. Click OK.
The Queries tool adds the comparison to the list on the Conditions tab.
Editing a Comparison
Removing a Comparison
- 141 -
1. In the Hub Console, start the Queries tool according to the instructions in
"Starting the Queries Tool" on page 129.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click the Conditions tab.
4. Select the comparison that you want to remove.
The Sort By tab displays a list of column(s) containing the values that the
query will use to sort the query results at run time. The information in this tab
corresponds to the SQL ORDER BY clause.
- 142 -
5. Expand the list for the table containing the column(s) that you want to
select for sorting.
The Queries tool displays the list of columns for the selected table.
- 143 -
Removing Table Column(s) from a Sort Order
To change the order in which the columns will appear in the result set (if the
list contains multiple columns):
1. In the Hub Console, start the Queries tool according to the instructions in
"Starting the Queries Tool" on page 129.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click the Sort tab.
4. Select one column that you want to move.
5. Do one of the following:
The Grouping tab displays a list of column(s) containing the values that the
query will use for grouping the query results at run time. The information in
this tab corresponds to the SQL GROUP BY clause.
- 144 -
Selecting the Grouping Columns
5. Expand the list for the table containing the column(s) that you want to
select for grouping.
The Queries tool displays the list of columns for the selected table.
- 145 -
6. Select the column(s) you want to use for grouping.
7. Click OK.
The Queries tool adds the selected column(s) to the list of columns on the
Grouping tab.
- 146 -
Changing the Column Order
To change the order in which the columns will be grouped in the result set (if
the list contains multiple columns):
1. In the Hub Console, start the Queries tool according to the instructions in
"Starting the Queries Tool" on page 129.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click the Grouping tab.
4. Select one column that you want to move.
5. Do one of the following:
The SQL tab displays the SQL statement that corresponds to the query options
you have specified for the selected query.
A custom query is simply a query for which you supply the SQL statement
directly, rather than building it according to the instructions in "Configuring
- 147 -
Queries" on page 130. Custom queries can be used in packages and in the data
steward tools.
Property Description
Query name Descriptive name for this query.
Description Option description of this query.
Query Group Select the query group to which this query belongs.
7. Click Finish.
The Queries tool displays the newly-added custom query.
- 148 -
8. Click the Edit button next to the SQL field.
9. Enter the SQL query according to the syntax rules for your database
platform.
Once you have created a custom query, you can modify its properties to refine
the criteria it uses to retrieve data from the ORS.
- 149 -
3. In the navigation tree, select the custom query that you want to modify.
4. Edit the property settings that you want to change, clicking the Edit button
You delete a custom query in the same way in which you delete a regular
query. For more information, see "Removing Queries" on page 151.
- 150 -
To view the impact analysis of a query:
1. In the Hub Console, start the Queries tool according to the instructions in
"Starting the Queries Tool" on page 129.
2. Expand the query group associated with the query you want to select.
3. Right click the query and choose Impact Analysis from the pop-up menu.
4. The Queries tool displays the Impact Analysis dialog.
5. Expand the list next to a table to display the columns associated with the
query, if you want.
6. Click Close.
Removing Queries
If a query has multiple packages based on it, remove those packages first
before attempting to remove the query.
To remove a query:
1. In the Hub Console, start the Queries tool according to the instructions in
"Starting the Queries Tool" on page 129.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Expand the query group associated with the query you want to remove.
4. Select the query you want to remove.
5. Right click the query and choose Delete Query from the pop-up menu.
The Queries tool prompts you to confirm deletion.
6. Click Yes.
The Queries tool removes the query from the list.
Configuring Packages
This section describes how to create and modify PUT and display packages.
You use the Packages tool in the Hub Console to define packages.
- 151 -
About Packages
A package is a public view of one or more underlying tables in Informatica
MDM Hub. Packages represent subsets of the columns in those tables, along
with any other tables that are joined to the tables. A package is based on a
query. The underlying query can select a subset of records from the table or
from another package. For more information, see "Configuring Queries" on
page 127.
- 152 -
When to Create a Package
You must create a package if you want your Informatica MDM Hub
implementation to:
• Merge and update records in the Hub Store using the Merge Manager and
Data Manager tools. For more information, see the Informatica MDM Hub
Data Steward Guide.
• Allow an external application user to access Informatica MDM Hub
functionality using Services Integration Framework (SIF) requests. For
more information, see the Informatica MDM Hub Services Integration
Framework Guide.
In most cases, you create one set of packages for the Merge Manager and
Data Manager tools, and a different set of packages for external application
users.
PUT-enabled packages:
• cannot include joins to other tables
• cannot be based on system tables or other packages
• cannot be based on queries that have constant columns, aggregate
functions, or group by settings
- 153 -
The Packages tool is divided into two panes:
Pane Description
navigation pane Displays a hierarchical list of configured packages.
properties pane Displays the properties of the selected package.
Adding Packages
To add a new package:
1. In the Hub Console, start the Packages tool according to the instructions in
"Starting the Packages Tool" on page 153.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Right-click in the Packages pane and choose New Package.
The Packages tool displays the New Package Wizard.
Note: If the welcome screen is displayed, click Next.
Field Description
Display Name of this package as it will be displayed in the Hub
Name Console.
Physical Actual name of the package in the database. Informatica MDM
Name Hub will suggest a physical name for the package based on the
display name that you enter.
Description Description of this package.
Enable PUT To create a PUT package, check (select) to insert or update
records into base object tables.
Note: Every package that you use for merging data or
updating data must be PUT-enabled.
If you do not enable PUT, you create a display (read-only)
package.
Secure Check (enable) to make this package a secure resource, which
Resource allows you to control access to this package. Once a package
- 154 -
Field Description
is designated as a secure resource, you can assign privileges
to it in the Roles tool. For more information, see "Securing
Informatica MDM Hub Resources" on page 629, and "Assigning
Resource Privileges to Roles" on page 641.
5. Click Next.
The New Package Wizard displays the Select Query dialog.
6. If you want, click New Query Group to add a new query group, as
described in "Configuring Query Groups" on page 129.
7. If you want, click New Query to add a new query, as described in
"Configuring Queries" on page 130.
8. Select a query.
Note: For PUT-enabled packages:
• only queries with ROWID_OBJECT can be used
• custom queries cannot be used
9. Click Finish.
The Packages tool adds the news package to the list.
- 155 -
5. If you want, expand the package in the packages list.
6. To change the query, select Query beneath the package and modify the
query as described in "Editing Query Properties" on page 132.
- 156 -
Refreshing Packages After Changing Queries
If a query has been changed, then any packages based on that query must be
refreshed.
To refresh a package:
1. In the Hub Console, start the Packages tool according to the instructions in
"Starting the Packages Tool" on page 153.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Select the package that you want to refresh.
4. From the Packages menu, choose Refresh.
Note: If after a refresh the query remains out of synch with the package, then
simply check (select) or uncheck (clear) any columns for this query. For more
information, see "Configuring the Column(s) in a Query" on page 136.
- 157 -
To expose this information:
1. Create a PUT-enabled base object package.
2. Create a query to join the PUT-enabled base object package with the other
tables.
3. Create a display package based on the query you just created.
Removing Packages
To remove a package:
1. In the Hub Console, start the Packages tool according to the instructions in
"Starting the Packages Tool" on page 153.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Select the package to remove.
4. Right click the package and choose Delete Package.
The Packages tool prompts you to confirm deletion.
5. Click Yes.
The Packages tool removes the package from the list.
- 158 -
Chapter 7: State Management
Chapter Contents
• "Before You Begin" on page 159
• "About State Management in Informatica MDM Hub" on page 159
• "State Transition Rules for State Management" on page 161
• "Configuring State Management for Base Objects" on page 162
• "Modifying the State of Records" on page 164
• "Rules for Loading Data" on page 168
- 159 -
System States
System state describes how base object records are supported by Informatica
MDM Hub. The following table describes the supported system states.
System States
State Description
ACTIVE Default state. Record has been reviewed and approved. Active
records participate in Hub processes by default.
This is a state associated with a base object or cross reference
record. A base object record is active if at least one of its cross
reference records is active. A cross reference record contributes to
the consolidated base object only if it is active.
These are the records that are available to participate in any
operation. If records are required to go through an approval
process, then these records have been through that process and
have been approved.
Note that Informatica MDM Hub allows matches to and from
PENDING and ACTIVE records.
PENDING Pending records are records that have not yet been approved for
general usage in the Hub. These records can have most operations
performed on them, but operations have to specifically request
pending records. If records are required to go through an approval
process, then these records have not yet been approved and are in
the midst of an approval process.
If there are only pending cross-reference records, then the Best
Version of the Truth (BVT) on the base object is determined through
trust on the PENDING records.
Note that Informatica MDM Hub allows matches to and from
PENDING and ACTIVE records.
DELETED Deleted records are records that are no longer desired to be part of
the Hub’s data. These records are not used in processes (unless
specifically requested). Records can only be deleted explicitly and
once deleted can be restored if desired. When a record that is
pending is deleted, it is physically deleted, does not enter the
DELETED state, and cannot be restored.
Note that Informatica MDM Hub does not include records in the
DELETED state for trust and validation rules.
- 160 -
Protecting Pending Records Using the Interaction
ID
You can not use the tools in the Hub Console to change the state of a base
object or cross-reference record from PENDING to ACTIVE state if the
interaction_ID is set. The Interaction ID column is used to protect a pending
cross-reference record from updates that are not part of the same process as
the original cross-reference record. Use one of the state management SIF API
requests, instead. For more information, see the .
Note: The Interaction ID can be specified through any API. However, it cannot
be specified when performing batch processing. For example, records that are
protected by an Interaction ID cannot be updated by the Load batch process.
State transition rules determine whether and when a record can change from
one state to another. State transition for base object and cross-reference
records can be enabled using the following methods:
• Using the Data Manager or Merge Manager tools in the Hub Console, as
described in the Informatica MDM Hub Data Steward Guide.
• Promote batch job (see "Promote Jobs" on page 552)
• SiperianClient API (see the Informatica MDM Hub Services Integration
Framework Guide)
State transition rules differ for base object and cross-reference records.
- 161 -
State Description
• Can transition to PENDING state only if the base object record
becomes DELETED and a pending cross-reference record is
added.
PENDING • Can transition to ACTIVE state. This transition is called
promotion. For more information, see "Modifying the State of
Records" on page 164.
• Cannot transition to DELETED state. Instead, a PENDING record
is physically removed from the Hub.
DELETED • Can transition to ACTIVE state only if cross-reference records
are restored.
• Cannot transition to PENDING state.
- 162 -
• Enable message queue triggers for a state-enabled base object record
(see "Enabling Message Queue Triggers for State Changes" on page 164)
Note: If the base object has custom query, when you disable state
management on the base object, you will always get a warning pop-up
window, even when the hub_state_ind is not included in the custom query.
- 163 -
3. Click the Enable State Management checkbox on the Advanced tab of
the Base Object properties.
4. Select Match/Merge Setup for the base object.
5. Click the Enable Match on Pending Records checkbox on the
Properties tab of Match/Merge Setup Details panel.
Informatica MDM Hub enables you to trigger message events for base object
record when a pending update occurs. The following message triggers are
available for state changes to base object or cross-reference records:
To enable the message queue triggers on a pending update for a base object:
1. Open the Model workbench and click on Schema.
2. In the Schema tool, click the Trigger on Pending Updates checkbox for
message queues in the Message Queues tool.
For more information about message queues and message triggers, including
how to enable message queue triggers for state changes to base object and
cross-reference records, see "Configuring Message Triggers" on page 456.
- 164 -
can set a record for promotion immediately using the Data Steward tools, or
you can flag records to be promoted at a later time using the Promote batch
process.
- 165 -
You can now promote these PENDING cross-reference records using the
Promote batch job.
To set up a batch job using the Batch Viewer to promote records flagged for
promotion:
1. Flag the desired PENDING records for promotion.
For more information, see "Modifying the State of Records" on page 164.
2. Open the Utilities workbench and click on the Batch Viewer tool.
3. Click on the Promote batch job under the Base Object node displayed in
the Batch Viewer.
4. Select Promote flagged records abc.
Where abc represents the associated records that you have previously
flagged for promotion.
5. Click Execute Batch button to promote the records flagged for
promotion.
To add a Promote Batch job using the Batch Group Tool to promote records
flagged for promotion:
1. Flag the desired PENDING records for promotion.
For more information, see "Modifying the State of Records" on page 164.
2. Open the Utilities workbench and click on the Batch Group tool.
- 166 -
3. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
4. Right-click the Batch Groups node in the Batch Group tree and choose Add
Batch Group from the pop-up menu (or select Add Batch Group from
the Batch Group menu). For more information, see "Adding Batch Groups"
on page 514.
5. In the batch groups tree, right click on any level, and choose the desired
option to add a new level to the batch group.
The Batch Group tool displays the Choose Jobs to Add to Batch Group
dialog. For more information, see "Adding Levels to a Batch Group" on
page 516.
6. Expand the base object(s) for the job(s) that you want to add.
7. Select the Promote flagged records in [XREF table] job.
8. Click OK.
The Batch Group tool adds the selected job(s) to the batch group.
- 167 -
You can now execute the batch group job. For more information, see
"Executing a Batch Group" on page 524.
The following table describes how input states affect the states of existing
cross-reference records.
Existing ACTIVE PENDING DELETED No XREF No Base
XREF (Load by Object
State: rowid) Record
Incoming
XREF
State:
ACTIVE Update Update + Update + Insert Insert
Promote Restore
PENDING Pending Pending Pending Pending Pending
Update Update Update + Update Insert
Restore
DELETED Soft Hard Hard Delete Error Error
Delete Delete
Undefined Treat Treat as Treat as Treat as Treat as
as PENDING DELETED ACTIVE ACTIVE
ACTIVE
- 168 -
Chapter 8: Configuring Hierarchies
Chapter Contents
• "About Configuring Hierarchies" on page 169
• "Starting the Hierarchies Tool" on page 178
• "Configuring Hierarchies" on page 191
• "Configuring Relationship Base Objects and Relationship Types" on page
193
• "Configuring Packages for Use by HM" on page 205
• "Configuring Profiles" on page 211
When you have finished defining Hierarchy Manager components, you can use
the package or query manager tools to update the query criteria.
Note: Packages have to be configured for use in HM as well, and the profile
has to be validated.
To understand the concepts in this chapter, you must be familiar with the
concepts in the following chapters in this guide:
• "Building the Schema" on page 73
• "Configuring Queries and Packages" on page 127
• "Configuring the Consolidate Process" on page 443
• "Setting Up Security" on page 621
- 169 -
Before You Begin
Before you begin to configure your Hierarchy Manager (HM) system, you must
have completed the following tasks:
• Start with a blank ORS or a valid ORS and register the database in CMX_
SYSTEM, as described in "Registering an ORS" on page 56.
• Verify that you have a license for Hierarchy Manager. For details, consult
your Informatica MDM Hub administrator.
• Perform data analysis, as described in "Preparing Your Data for Hierarchy
Manager" on page 170.
Note: The same options you see on the right-click menu in the Hierarchy
Manager are also available on the Hierarchies menu.
- 170 -
For more information on security issues, see "Setting Up Security" on page
621.
• Created valid schema to work with Informatica MDM Hub and the HM.
For more information on schemas and how to create them, see "Building
the Schema" on page 73.
• Created all relationships between your entities, including:
• Hierarchical relationships:
• All child entities must have a valid parent entity related to them.
Your data cannot have any ‘orphan’ child entities when it enters
HM.
• All hierarchies must be validated (see "Informatica MDM Hub
Processes" on page 218).
• Foreign key relationships.
For a general overview of foreign key relationships, see "Process
Overview for Defining Foreign-Key Relationships" on page 114. For
more information about parent-child relationships, see "Configuring
Match Paths for Related Records" on page 373.
• One-hop and multi-hop relationships (direct and indirect relationships
between entities). For more information on these kinds of
relationships, see the Informatica MDM Hub Data Steward Guide.
• Derived HM types.
• Consolidated duplicate entities from multiple source systems.
For example, a group of entities (Source A) might be the same as another
group of entities (Source B), but the two groups of entities might have
different group names. Once the entities are identified as being identical,
the two groups can be consolidated.
For more information on consolidation, see "Informatica MDM Hub
Processes" on page 218.
• Grouped your entities into logical categories, such as physician’s names
into the “Physician” category.
For more information on how to group your data, see "Configuring
Operational Reference Stores and Datasources" on page 54.
• Made sure that your data complies with the rules for:
• Referential integrity.
• Invalid data.
• Data volatility.
For more information on these database concepts, see a database
reference text.
- 171 -
Use Case Example of How to Prepare Data for
Hierarchy Manager
This section contains an example of how to manipulate your data before it
enters Informatica MDM Hub and before it is viewed in Hierarchy Manager.
Typically, a company’s data would be much larger than the example given
here.
Scenario
John has been tasked with manipulating his company’s data so that it can be
viewed and used within Hierarchy Manager in the most efficient way. To
simplify the example, we are describing a subset of the data that involves
product types and products of the company, which sells computer
components.
The company sells three types of products: mice, trackballs, and keyboards.
Each of these product types includes several vendors and different levels of
products, such as the Gaming keyboard and the TrackMan trackball.
Methodology
In this step you organize the data into the Hierarchy that will then be
translated into the HM configuration.
The following table (which contains data from the Marketing department)
shows an example of how John might organize his data.
- 172 -
Note: Most data sets will have many more items.
The table shows the data that will be stored in the Products BO. This is the BO
to convert (or create) in HM. The table shows Entities, such as Mice or Laser
Mouse. The relationships are shown by the grouping, that is, there is a
relationship between Mice and Laser Mouse. The heading values are the Entity
Types: Mice is a Product Group and Laser Mouse is a Product. This Type is
stored in a field on the Product table.
Organizing the data in this manner allows John to clearly see how many
entities and entity types are part of the data, and what relationships those
entities have.
The major category is ProdGroup, which can include both a product group
(such as mice and pointers), the category Product, and the products
themselves (such as the Trackman Wheel). The relationships between these
items can be encapsulated in a relationship object, which John calls Product
Rel. In the information for the Product Rel, John has explained the
relationships: Product Group is the parent of both Product and Product Group.
John begins by accessing the Hierarchy Tool. When he accesses the tool, the
system creates the Relationship Base Object Tables (RBO tables). RBO tables
are essentially system base objects that are required base objects containing
specific columns. They store the HM configuration data, such as the data that
you see in the table in Step 1.
For instructions on how to create base objects, see "Configuring Base Objects"
on page 82. This section describes the choices you would make when you
create the example base objects in the Schema tool.
- 173 -
You must create and configure a base object for each entity object and
relationship object that you identified in the previous step. In the example,
you would create a base object for Product and convert it to an HM Entity
Object. The Product Rel BO should be created in HM directly (an easier
process) instead of converting. Each new base object is displayed in the
Schema panel under the category Base Objects. Repeat this process to create
all your base objects.
In the next section, you configure the base objects so that they are optimized
for HM use.
You created the two base objects (Product and Product Rel) in the previous
section. This section describes how to configure them.
Configuring a base object involves filling in the criteria for the object’s
properties, such as the number and type of columns, the content of the staging
tables, the name of the cross-reference tables (if any), and so on. You might
also enable the history function, set up validation rules and message triggers,
create a custom index, and configure the external match table (if any).
Whether or not you choose these options and how you configure them depends
on your data model and base object choices.
In the example, John configures his base objects as the following sections
explain.
Note: Not all components of the base-object creation are addressed here,
only the ones that have specific significance for data that will be used in the
HM. For more information on the components not discussed here, see
"Building the Schema" on page 73.
Columns
- 174 -
This table shows the Product BO after conversion to an HM entity object. In
this list, only the Product Type field is an HM field.
Every base object has system columns and user-defined columns. System
columns are created automatically, and include the required column: Rowid
Object. This is the Primary key for each base object table and contains a
unique, Hub-generated value. This value cannot be null because it is the HM
lookup for the class code. HM makes a foreign key constraint in the database
so a ROWID_OBJECT value is required and cannot be null.
For the user-defined columns, John choose logical names that would
effectively include information about the products, such as Product Number,
Product Type, and Product Description. These same column and column values
must appear in the staging tables.
Staging Tables
John makes sure that all the user-defined columns from the staging tables are
added as columns in the base object, as the graphic above shows. The Lookup
column shows the HM-added lookup value.
Notice that several columns in the Staging Table (Status Cd, Product Type, and
Product Type Cd) have references to lookup tables. You can set these
references up when you create the Staging Table. You would use lookups if
you do not want to hardcode a value in your staging table, but would rather
have the server look up a value in the parent table.
Most of the lookups are unrelated to HM and are part of the data model. The
Rbo Bo Class lookup is the exception because it was added by HM. HM adds
the lookup on the product Type column.
Note: When you are converting entities to entity base objects (entities that
are configured to be used in HM), you must have lookup tables to check the
values for the Status Cd, Product Type, and Product Type Cd.
Warning: HM Entity objects do not require start and end dates. Any start and
end dates would be user defined. However, Rel Objects do use these. Do not
create new Rel Objects with different names for start and end dates. These
are already provided.
- 175 -
Step 4 - Creating Entity Types
You create entity types in the Hierarchy Tool. John creates two entity types:
ProdGroup and Product Type. The following figure shows the completed
Product Entity Type information.
Each entity type has a code that derives from the data analysis and the design.
In this example, John chose to use Product as one type, and Product Group as
another.
This code must be referenced in the corresponding RBO base object table. In
this example, the code Product is referenced in the C_RBO_BO_CLASS table.
The value of the BO_CLASS_CODE is ‘Product’.
The following figure shows the relationship between the HM entity objects and
HM relationship objects to the RBO tables:
- 176 -
When John has completed all the steps in this section, he will be ready to
create other HM components, such as packages, and to view his data in the
HM. For example, the following graphic shows the relationships that John has
set up in the Hierarchies Tool, displayed in the Hierarchy Manager. This
example shows the hierarchy involving Mice devices fully. For more
information on how to use HM, see the Informatica MDM Hub Data Steward
Guide.
- 177 -
Starting the Hierarchies Tool
To start the Hierarchies tool:
• In the Hub Console, do one of the following:
• Expand the Model workbench, and then click Hierarchies.
OR
If you are setting up the Hierarchies tool, see "Creating the HM Repository
Base Objects" on page 178. If you already have RBO tables set up, see
"Configuring Entity Icons" on page 180.
Queries and MRM packages (and their associated queries) will also be created
for these RBO tables.
Note: Any option that you can select by right-clicking in the navigation panel,
you can also choose from the Hierarchies tool menu.
After you start the Hierarchies tool, if an ORS does not have the necessary
RBO tables, then the Hierarchies tool walks you through the process of
creating them.
The following steps explain what to select in the dialog boxes that the
Hierarchies tool displays:
- 178 -
1. Choose Yes in the Hub Console dialog to create the metadata (RBO tables)
for HM in the ORS.
2. Select the tablespace names in the Create RBO tables dialog, and then
click OK.
- 179 -
3. Launch the Hierarchies tool in the Hub Console.
4. Click Yes to add additional columns.
After you upgrade a pre-XU schema to XU, you will be reminded to remove
obsolete HM metadata when you get into the Hierarchies tool.
Note: If the Rbo Rel Type Usage base object is being used by some other non-
HM base object, you will be told to manually delete the table by going to the
schema manager.
Informatica MDM Hub shows relationship and entity types under the base
object with which they are associated. If a type is not associated with a base
object, for example it does not have packages assigned, it is not displayed in
the GUI, but does remain in the database.
During the ORS upgrade process, the migration script skips over the orphan
entity and relationship types, displays a related warning message, then
continues. After the ORS upgrade, you can delete the orphan types or
associate entities and relationship types with them.
If you want to associate orphan types but you have not created the
corresponding base objects, create the objects, then press refresh. The
software prompts you to create the association.
To import your own icons, create a ZIP or JAR file containing your icons. For
each icon, create a 16 x 16 icon for the small icon and a 48 x 48 icon for the
large icon.
- 180 -
1. Acquire a write lock.
2. Start the Hierarchies tool.
3. Right-click anywhere in the navigation pane and choose Add Entity
Icons.
Note: You must acquire a lock to display the right-click menu.
A browse files window opens.
4. Browse for the JAR or ZIP file containing your icons.
5. Click Open to add the icons.
You cannot modify icons directly from the console. You can download a ZIP or
JAR file, modify its contents, then upload it again into the console.
You can either delete icons groups or make them inactive. If an icon is already
associated with an entity, or if you could use a group of icons in the future,
you might consider choosing to inactivate them instead of deleting them.
Warning: Informatica MDM Hub does not validate icons assignments before
deleting. If you delete an icon that is currently assigned to an Entity Type, you
will get an error when you try to save the edit.
You cannot delete individual icons from a ZIP or JAR file from the console; you
can only delete them as a group or package.
- 181 -
About Entities, Entity Objects, and Entity Types
This section describes entities, entity objects, and entity types in Hierarchy
Manager.
Entities
An entity base object is a base object that has been configured in HM, and that
is used to store HM entities. When you create an entity base object using the
Hierarchies tool (instead of the Schema Manager), the Hierarchies tool
automatically creates the columns required for Hierarchy Manager. You can
also convert an existing MRM base object to an entity base object by using the
options in the Hierarchies tool.
After adding an entity base object, you use the Schema Manager to view, edit,
or delete it. For more information, see "Configuring Base Objects" on page 82.
Entity Types
- 182 -
• A more general entity type, such as “care provider” (which encompasses
nurses, nurse practitioners, doctors, and others) is not granular enough. In
this case, the types of relationships that such a general entity type will
have will depend on something beyond just the entity type. Therefore, you
need to need to define more-granular entity types.
4. Click OK.
The Hierarchies tool prompts you to enter information about the new base
object.
Field Description
Item Type Read-only. Already specified.
Display Name of this base object as it will be displayed in the Hub
- 183 -
Field Description
name Console.
Physical Actual name of the table in the database. Informatica MDM
name Hub will suggest a physical name for the table based on the
display name that you enter.
The RowId is generated and assigned by the system, but the
BO Class Code is created by the user, making it easier to
remember.
Data Name of the data tablespace. For more information, see the
tablespace Informatica MDM Hub Installation Guide.
Index Name of the index tablespace. For more information, see the
tablespace Informatica MDM Hub Installation Guide.
Description Description of this base object.
Foreign Column used as the Foreign Key for this entity type; can be
Key either ROWID or CODE.
column for
Entity The ability to choose a BO Class CODE column reduces the
Types complexity by allowing you to define the foreign key
relationship based on a predefined code, rather than the
Informatica MDM Hub-generated ROWID.
Display Descriptive name of the column of the Entity Type Foreign Key
name that is displayed in Hierarchy Manager.
Physical Actual name of the FK column in the table. Informatica MDM
name Hub will suggest a physical name for the FK column based on
the display name that you enter.
6. Click OK to save the new base object.
The base object you created has the columns required by Hierarchy Manager.
You probably require additional columns in the base object, which you can add
using the Schema Manager, as described in "Configuring Columns in Tables"
on page 102.
Important: When you modify the base object using the Schema Manager, do
not change any of the columns added by Hierarchy Manager. Modifying any of
these Hierarchy Manager columns will result in unpredictable behavior and
possible data loss.
You must convert base objects to entity base objects before you can use them
in HM.
Base objects created in MRM do not have the metadata required by Hierarchy
Manager. In order to use these MRM base objects with HM, you must add this
metadata via a conversion process. Once you have done this, you can use
these converted base objects with both MRM and HM.
- 184 -
2. Right-click anywhere in the navigation pane and choose Convert BO to
Entity/Relationship Object.
Note: The same options you see on the right-click menu are also available
on the Hierarchies menu.
3. In the Modify Existing Base Object dialog, select Convert to Entity and
click OK.
Note: If you do not see any choices in the Modify Base Object field, then
there are no non-hierarchy base objects available. You must create one in
the Schema tool.
4. Click OK.
If the base object already has HM metadata, the Hierarchies tool will
display a message indicating the HM metadata that exists.
5. In the Foreign Key Column for Entity Types field, select the column to be
added: RowId Object or BO Class Code.
- 185 -
This is the descriptive name of the column of the Entity Type Foreign Key
that is displayed in Hierarchy Manager.
The ability to choose a BO Class Code column reduces the complexity by
allowing you to define the foreign key relationship based on a predefined
code, rather than the Informatica MDM Hub-generated ROWID.
6. In the Existing BO Column to use, select an existing column or select the
Create New Column option.
If no BO columns exist, only the Create New Column option is available.
7. In the Display Name and Physical Name fields, create display and physical
names for the column, and click OK.
The base object will now have the columns that Hierarchy Manager requires.
To add additional columns, use the Schema Manager (see "Configuring
Columns in Tables" on page 102).
Important: When you modify the base object using the Schema Manager
tool, do not change any of the columns added using the Hierarchies tool.
Modifying any of these columns will result in unpredictable behavior and
possible data loss.
The Hierarchies tool displays a new entity type (called New Entity Type)
in the navigation tree under the Entity Object you selected.
- 186 -
2. In the properties panel, specify the following properties for this new entity
base object.
Field Description
Code Unique code name of the Entity Type. Can be used as a foreign
key from HM entity base objects.
Display Name of this entity type as it will be displayed in the Hub
name Console. Specify a unique, descriptive name.
Description Description of this entity type.
Color Color of the entities associated with this entity type as they
will be displayed in the Hub Console in the Hierarchy Manager
Console and Business Data Director.
Small Icon Small icon for entities associated with this entity type as they
will be displayed in the Hub Console in the Hierarchy Manager
Console and Business Data Director.
Large Icon Large icon for entities associated with this entity type as they
will be displayed in the Hub Console in the Hierarchy Manager
Console and Business Data Director.
The color you choose determines how entities of this type are displayed in
the Hierarchy Manager. Select a color and click OK.
- 187 -
4. To select a small icon for the new entity type, click next to Small
Icon.
The Choose Small Icon window is displayed.
Small icons determine how entities of this type are displayed when the
graphic view window shows many entities. To learn more about adding
icon graphics for your entity types, see "Configuring Entity Icons" on page
180.
Select a small icon and click OK.
5. To select a large icon for the new entity type, click next to Large Icon.
The Choose Large Icon window is displayed.
Large icons determine how entities of this type are displayed when the
graphic view window shows few entities. To learn more about adding icon
graphics for your entity types, see "Configuring Entity Icons" on page 180.
Select a large icon and click OK.
- 188 -
2. For each field that you want to edit, click and make the change that
you want. For more information about these fields, see "Adding Entity
Types" on page 186.
3. When you have finished making changes, click to save your changes.
Warning: If your entity object uses the code column, you probably do not
want to modify the entity type code if you already have records for that entity
type.
You can delete any entity type that is not used by any relationship types. If the
entity type is being used by one or more relationship types, attempting to
delete it will generate an error.
Warning: You probably do not want to delete an entity type if you already
have entity records that use that type. If your entity object uses the code
column instead of the rowid column and you have records in that entity object
for the entity type you are trying to delete, you will get an error.
In addition to configuring color and icons for entities, you can also configure
the font size and maximum width. While color and icons can be specified for
each entity type, the font size and width apply to entities of all types.
To change the font size in HM, use the HM Font Size and Entity Box Size. The
default entity font size (38 pts) and max entity box width (600 pixels) can be
overridden by settings in the cmxserver.properties file. The settings to use
are:
sip.hm.entity.font.size=fontSize
sip.hm.entity.max.width=maxWidth
- 189 -
The value for fontSize can be from 6 to 100 and the value for maxWidth can be
from 20 to 5000. If value specified is outside the range, the minimum or
maximum values are used. Default values are used if the values specified are
not numbers.
Note that when you revert the entity object, you are also reverting its
corresponding relationship objects.
4. In the Revert Entity/Relationship Object dialog box, click OK.
- 190 -
A dialog is displayed when the entity is reverted.
Configuring Hierarchies
This section describes how to define hierarchies using the Hierarchies tool.
About Hierarchies
A hierarchy is a set of relationship types (as described in "About Relationships,
Relationship Objects, and Relationship Types" on page 193). These
relationship types are not ranked, nor are they necessarily related to each
other. They are merely relationship types that are grouped together for ease
of classification and identification. The same relationship type can be
associated with multiple hierarchies. A hierarchy type is a logical classification
of hierarchies.
Adding Hierarchies
To add a new hierarchy:
1. In the Hierarchies tool, acquire a write lock.
2. Right-click an entity object in the navigation pane and choose Add
Hierarchy.
- 191 -
The Hierarchies tool displays a new hierarchy (called New Hierarchy) in
the navigation tree under the Hierarchies node. The default properties are
displayed in the properties pane.
Field Description
Code Unique code name of the hierarchy. Can be used as a foreign
key from HM relationship base objects.
Display Name of this hierarchy as it will be displayed in the Hub
name Console. Specify a unique, descriptive name.
Description Description of this hierarchy.
Editing Hierarchies
To edit a hierarchy:
1. In the Hierarchies tool, acquire a write lock.
2. In the navigation tree, click the hierarchy to edit.
Warning: If your relationship object uses the hierarchy code column (instead
of the rowid column), you probably do not want to modify the hierarchy code if
you already have records for that hierarchy in the relationship object.
Deleting Hierarchies
Warning: You do not want to delete a hierarchy if you already have
relationship records that use the hierarchy. If your relationship object uses
- 192 -
the hierarchy code column instead of the rowid column and you have records
in that relationship object for the hierarchy you are trying to delete, you will
get an error.
To delete a hierarchy:
1. In the Hierarchies tool, acquire a write lock.
2. In the navigation tree, right-click the hierarchy that you want to delete,
and choose Delete Hierarchy.
The Hierarchies tool prompts you to confirm deletion.
3. Choose Yes.
The Hierarchies tool removes the selected hierarchy from the list.
Note: You are allowed to delete a hierarchy that has relationship types
associated with it. There will be a warning with the list of associated
relationship types. If you elect to delete the hierarchy, all references to it will
automatically be removed.
Relationships
- 193 -
Relationship Types
A well defined set of Hierarchy Manager relationship types has the following
characteristics:
• It reflects the real-world relationships between your entity types.
• It supports multiple relationship types for each relationship.
- 194 -
The Hierarchies tool prompts you to enter information about the new
relationship base object.
5. Specify the following properties for this new entity base object.
Field Description
Item Type Read-only. Already specified.
Display Name of this base object as it will be displayed in the Hub
name Console.
Physical Actual name of the table in the database. Informatica MDM
name Hub will suggest a physical name for the table based on the
display name that you enter.
Data Name of the data tablespace. For more information, see the
tablespace Informatica MDM Hub Installation Guide.
Index Name of the index tablespace. For more information, see the
tablespace Informatica MDM Hub Installation Guide.
Description Description of this base object.
Entity Base Entity base object to be linked via this relationship base
Object 1 object.
Display Name of the column that is a FK to the entity base object 1.
name
Physical Actual name of the column in the database. Informatica MDM
name Hub will suggest a physical name for the column based on the
display name that you enter.
Entity Base Entity base object to be linked via this relationship base
Object 2 object.
- 195 -
Field Description
Display Name of the column that is a FK to the entity base object 2.
name
Physical Actual name of the column in the database. Informatica MDM
name Hub will suggest a physical name for the column based on the
display name that you enter.
Hierarchy Column used as the foreign key for the hierarchy; can be
FK Column either ROWID or CODE.
The ability to choose a BO Class CODE column reduces the
complexity by allowing you to define the foreign key
relationship based on a predefined code, rather than the
Informatica MDM Hub-generated ROWID.
Hierarchy Name of this FK column as it will be displayed in the Hub
FK Display Console
Name
Hierarchy Actual name of the hierarchy foreign key column in the table.
FK Physical Informatica MDM Hub will suggest a physical name for the
Name column based on the display name that you enter.
Rel Type Column used as the foreign key for the relationship; can be
FK Column either ROWID or CODE.
Rel Type Name of the column that is used to store the Rel Type CODE or
Display ROWID.
Name
Rel Type Actual name of the relationship type FK column in the table.
Physical Informatica MDM Hub will suggest a physical name for the
Name column based on the display name that you enter.
6. Click OK to save the new base object.
The relationship base object you created has the columns required by
Hierarchy Manager. You may require additional columns in the base object,
which you can add using the Schema Manager, as described in "Configuring
Columns in Tables" on page 102.
Important: When you modify the base object using the Schema Manager, do
not change any of the columns added by Hierarchy Manager. Modifying any of
these columns will result in unpredictable behavior and possible data loss.
A foreign key relationship base object is an entity base object with a foreign
key to another entity base object.
- 196 -
3. Specify the base object and the number of Foreign Key columns, then click
OK.
The Hierarchies tool displays the Convert to FK Relationship Base Object
dialog.
Field Description
FK Select FK entity base object from list.
Constraint
Entity BO 1
Existing BO Name of existing base object column used for FK, or choose
Column to to create a new column.
Use
FK Column Name of FK column as it will be displayed in the Hub Console.
Display
Name 1
FK Column Actual name of FK column in the database. Informatica MDM
Physical Hub will suggest a physical name for the table based on the
Name 1 display name that you enter.
FK Column Choose Entity1 or Entity2, depending on what the FK Column
Represents represents in the relationship.
- 197 -
5. Click OK to save the new FK relationship object.
The base object you created has the columns required by Hierarchy Manager.
You may require additional columns in the base object, which you can add
using the Schema Manager, as described in "Configuring Columns in Tables"
on page 102.
Important: When you modify the base object using the Schema Manager
tool, do not change any of the columns added by the Hierarchies tool.
Modifying any of these columns will result in unpredictable behavior and
possible data loss.
For more information about foreign key relationships, see "Building the
Schema" on page 73
Relationship base objects are tables that contain information about two entity
base objects.
Base objects created in MRM do not have the metadata required by Hierarchy
Manager for relationship information. In order to use these MRM base objects
with Hierarchy Manager, you must add this metadata via a conversion
process. Once you have done this, you can use these converted base objects
with both MRM and HM.
3. Click OK.
The Convert to Relationship Base Object screen is displayed.
- 198 -
4. Specify the following properties for this base object.
Field Description
Entity Entity base object to be linked via this relationship base object.
Base
Object 1
Display Name of the column that is a FK to the entity base object 1.
name
Physical Actual name of the column in the database. Informatica MDM
name Hub will suggest a physical name for the column based on the
display name that you enter.
Entity Entity base object to be linked via this relationship base object.
Base
Object 2
Display Name of the column that is a FK to the entity base object 2.
name
Physical Actual name of the column in the database. Informatica MDM
name Hub will suggest a physical name for the column based on the
display name that you enter.
- 199 -
Field Description
Hierarchy Column used as the foreign key for the hierarchy; can be either
FK ROWID or CODE.
Column The ability to choose a BO Class CODE column reduces the
complexity by allowing you to define the foreign key
relationship based on a predefined code, rather than the
Informatica MDM Hub-generated ROWID.
Existing Actual column in the existing BO to use.
BO
Column
to Use
Hierarchy Name of this FK column as it will be displayed in the Hub
FK Console
Display
Name
Hierarchy Actual name of the hierarchy foreign key column in the table.
FK Informatica MDM Hub will suggest a physical name for the
Physical column based on the display name that you enter.
Name
Rel Type Column used as the foreign key for the relationship; can be
FK either ROWID or CODE.
Column
Existing Actual column in the existing BO to use.
BO
Column
to Use
Rel Type Name of the FK column that is used to store the Rel Type CODE
FK or ROWID.
Display
Name
Rel Type Actual name of the relationship type FK column in the table.
FK Informatica MDM Hub will suggest a physical name for the
Physical column based on the display name that you enter.
Name
5. Click OK.
Warning: When you modify the base object using the Schema Manager tool,
do not change any of the columns added by HM. Modifying any of these HM
columns will result in unpredictable behavior and possible data loss.
- 200 -
4. A dialog is displayed when the entity is reverted.
- 201 -
Note: You can only save a relationship type if you associate it with a
hierarchy.
A Foreign Key Relationship Base Object is an Entity Base Object containing
a foreign key to another Entity Base Object. A Relationship Base Object is
a table that relates the two Entity Base Objects.
Note: FK relationship types can only be associated with a single hierarchy.
3. The properties panel displays the properties you must enter to create the
relationship.
4. In the properties panel, specify the following properties for this new
relationship type.
Field Description
Code Unique code name of the rel type. Can be used as a foreign
key from HM relationship base objects.
Display Name of this relationship type as it will be displayed in the
- 202 -
Field Description
name Hub Console. Specify a unique, descriptive name.
Description Description of this relationship type.
Color Color of the relationships associated with this relationship
type as they will be displayed in the Hub Console in the
Hierarchy Manager Console and Business Data Director.
Entity Type First entity type associated with this new relationship type.
1 Any entities of this type will be able to have relationships of
this relationship type.
Entity Type Second entity type associated with this new relationship type.
2 Any entities of this type will be able to have relationships of
this relationship type.
Direction Select a direction for the new relationship type to allow a
directed hierarchy. The possible directions are:
• Entity 1 to Entity 2
• Entity 2 to Entity 1
• Undirected
• Bi-Directional
• Unknown
An example of a directed hierarchy is an organizational chart,
with the relationship reports to being directed from employee
to supervisor, and so on, up to the head of the organization.
FK Rel The start date of the foreign key relationship.
Start Date
FK Rel End The end date of the foreign key relationship.
Date
Hierarchies Check the check box next to any hierarchy that you want
associated with this new relationship type. Any selected
hierarchies can contain relationships of this relationship type.
The color you choose determines how entities of this type are displayed in
the Hierarchy Manager. Select a color and click OK.
- 203 -
6. Click the Calendar button to designate a start and end date for a
foreign key relationship. All relationships of this FK relationship type will
have the same start and end date. If you do not specify these dates, the
default values are automatically added.
7. Select a hierarchy.
3. For each field that you want to edit, click and make the change that
you want. To learn more about these fields, see "Adding Relationship
Types" on page 201.
4. When you have finished making changes, click to save your changes.
Warning: If your relationship object uses the code column, you probably do
not want to modify the relationship type code if you already have records for
that relationship type.
The above warnings are not applicable to FK relationship types.You can delete
relationship types that are associated with hierarchies. The confirmation
dialog displays the hierarchies associated with the relationship type being
deleted.
- 204 -
The Hierarchies tool removes the selected relationship type from the list.
About Packages
As described in "Configuring Queries and Packages" on page 127, a package is
a public view of one or more underlying tables in Informatica MDM Hub.
Packages represent subsets of the columns in those tables, along with any
other tables that are joined to the tables. A package is based on a query. The
underlying query can select a subset of records from the table or from another
package. Packages are used for configuring user views of the underlying data.
For more information, see "Configuring Queries and Packages" on page 127.
You must first create a package to use with Hierarchy Manager, then you must
associate it with Entity Types or Relationship Types.
Creating Packages
This section describes how to create HM and Relationship packages.
To create an HM package:
1. Acquire a write lock.
2. In the Hierarchies tool, right-click anywhere in the navigation pane and
choose Create New Package.
The Hierarchies tool starts the Create New Package wizard and displays
the first dialog box.
- 205 -
3. Specify the following information for this new package.
Field Description
Type of One of the following types:
Package • Entity Object
• Relationship Object
• FK Relationship Object
Query Select an existing query group or choose to create a new one.
Group In Informatica MDM Hub, query groups are logical groups of
queries. For more information, see "Configuring Query
Groups" on page 129.
Query Name of the new query group - only needed if you chose to
group create a new group above.
name
Description Optional description for the new query group you are creating.
4. Click Next.
The Create New Package wizard displays the next dialog box.
- 206 -
5. Specify the following information for this new package.
Field Description
Query Name of the query. In Informatica MDM Hub, a query is a
Name request to retrieve data from the Hub Store. For more
information, see "Configuring Queries" on page 130.
Description Optional description.
Select Primary table for this query.
Primary
Table
6. Click Next.
The Create New Package wizard displays the next dialog box.
- 207 -
Field Description
Display Display name for this package, which will be used to display
Name this package in the Hub Console.
Physical Physical name for this package. The Hub Console will suggest
Name a physical name based on the display name you entered.
Description Optional description.
Enable PUT Select to enable records to be inserted or changed. (optional)
If you do not choose this, your package will be read only. If
you are creating a foreign key relationship object package,
you have additional steps in Step 9 of this procedure.
Note: You must have both a PUT and a non-PUT package for
every Foreign Key relationship. Both Put and non-Put packages
that you create for the same foreign key relationship object
must have the same columns.
Secure Select to create a secure resource. (optional)
Resource
8. Click Next.
The Create New Package wizard displays a final dialog box. The dialog box
you see depends on the type of package you are creating.
• If you selected to create either a package for entities or relationships
or a PUT package for FK relationships, a dialog box similar to the
following dialog box is displayed. The required columns (shown in
grey) are automatically selected — you cannot deselect them.
Deselect the columns that are irrelevant to your package.
Note: You must have both a PUT and a non-PUT package for every
Foreign Key relationship. Both Put and non-Put packages that you
create for the same foreign key relationship object must have the
same columns.
- 208 -
• If you selected to create a non-Put enabled package for foreign key
relationships (see Step 7 of this procedure - do not check the Put check
box), the following dialog box is displayed.
Use the Packages tool to view, edit, or delete this newly-created package, as
described in "Configuring Packages" on page 151.
You should not remove columns that are needed by Hierarchy Manager. These
columns are automatically selected (and greyed out) when the user creates
packages using the Hierarchies tool.
- 209 -
Assigning Packages to Entity or Relationship Types
After you create a profile, and a package for each of the entity/relationship
types in a profile, you must assign the packages. This defines what fields are
displayed when an entity is displayed in HM. For more information, see
"Customizing the Hub Console Interface" on page 45. You can also assign a
package for relationship types and entity types.
The numbers in the cells define the sequence in which the attributes are
displayed.
3. Configure the package for your entity or relationship type.
Field Description
Label Columns used to display the label of the entity/relationship you
are viewing in the HM graphical console. These columns are used
to create the Label Pattern in the Hierarchy Manager Console and
Business Data Director.
To edit a label, click the label value to the right of the label. In
the Edit Pattern dialog, enter a new label or double-click a
column to use it in a pattern.
Tooltip Columns used to display the description or comment that
appears when you scroll over the entity/relationship. Used to
create the tooltip pattern in the Hierarchy Manager Console and
Business Data Director.
To edit a tooltip, click the tooltip pattern value to the right of the
Tooltip Pattern label. In the Edit Pattern dialog, enter a new
tooltip pattern or double-click a column to use it in a pattern.
- 210 -
Field Description
Common Columns used when entities/relationships of different types are
displayed in the same list. The selected columns must be in
packages associated with all Entity/Relationship Types in the
Profile.
Search Columns that can be used with the search tool.
List Columns to be displayed in a search result.
Detail Columns used for the detailed view of an entity/relationship
displayed at the bottom of the screen.
Put Columns that are displayed when you want to edit a record.
Add Columns that are displayed when you want to create a new
record.
4. When you have finished making changes, click to save your changes.
Configuring Profiles
This section describes how to configure profiles using the Hierarchies tool.
About Profiles
In Hierarchy Manager, a profile is used to define user access to HM objects—
what users can view and what the HM objects look like to those users. A
profile determines what fields and records an HM user may display, edit, or
add. For example, one profile can allow full read/write access to all entities
and relationships, while another profile can be read-only (no add or edit
operations allowed). Once you define a profile, you can configure it as a
secure resource, as described in "Securing Informatica MDM Hub Resources"
on page 629.
Adding Profiles
A new profile (called Default) is created automatically for you before you
access the HM. The default profile can be maintained, and you can also add
additional profiles.
Note: The Business Data Director uses the Default Profile to define how Entity
Labels as well as Relationship and Entity Tooltips are displayed. Additional
Profiles, as well as the additional information defined within Profiles, is only
used within the Hierarchy Manager Console and not the Business Data
Director.
- 211 -
The Hierarchies tool displays a new profile (called New Profile) in the
navigation tree under the Profiles node. The default properties are
displayed in the properties pane.
When you select these relationship types and click Save, the tree below
the Profile will be populated with Entity Objects, Entity Types, Rel Objects
and Rel Types. When you deselect a Rel type, only the Rel types will be
removed from the tree - not the Entity Types.
3. Specify the following information for this new profile.
Field Description
Name Unique, descriptive name for this profile.
Description Description of this profile.
Relationship Select one or more relationship types associated with
Types this profile.
Editing Profiles
To edit a profile:
1. Acquire a write lock.
2. In the Hierarchies tool, in the navigation tree, click the profile that you
want to edit.
- 212 -
3. Configure the profile as needed (specifying the appropriate profile name,
description, and relationship types and assigning packages), according to
the instructions in "Adding Profiles" on page 211 and "Configuring
Packages for Use by HM" on page 205.
4. When you have finished making changes, click to save your changes.
Validating Profiles
To validate a profile:
1. Acquire a write lock.
2. In the Hierarchies tool, in the navigation pane, select the profile to
validate.
- 213 -
4. Select a sandbox to use.
For information about creating and configuring sandboxes, see the
Informatica MDM Hub Data Steward Guide.
5. To validate the data, check Validate Data. This may take a long time if
you have a lot of records.
6. To start the validation process, click Validate HM Configuration.
The Hierarchies tool displays a progress window during the validation
process. The results of the validation appear in the window below the
buttons.
- 214 -
8. Choose the directory where the validation report will be saved.
9. Click Clear to clear the box containing the description of the validation
results.
Copying Profiles
To copy a profile:
1. Acquire a write lock.
2. In the Hierarchies tool, right-click the profile that you want to copy, and
then choose Copy Profile.
The Hierarchies tool displays a new profile (called New Profile) in the
navigation tree under the Profiles node. This new profile that is an exact
copy (with a different name) of the profile that you selected to copy. The
default properties are displayed in the properties pane.
Deleting Profiles
To delete a profile:
1. Acquire a write lock.
2. In the Hierarchies tool, right-click the profile that you want to delete, and
choose Delete Profile.
The Hierarchies tool displays a window that warns that packages will be
removed when you delete this profile.
3. Click Yes.
The Hierarchies tool removes the deleted profile.
- 215 -
1. Acquire a write lock.
2. In the Hierarchy tool, right-click the relationship type and choose Delete
Entity Type/Relationship Type From Profile.
If the profile contains relationship types that use the entity/relationship
type that you want to delete, you will not be able to delete it unless you
delete the relationship type from the profile first.
Sandboxes
To learn about sandboxes, see the Hierarchy Manager chapter in the
Informatica MDM Hub Data Steward Guide.
- 216 -
Part 3: Configuring the Data Flow
Contents
• "Informatica MDM Hub Processes" on page 218
• "Configuring the Land Process" on page 264
• "Configuring the Stage Process" on page 274
• "Configuring Data Cleansing" on page 307
• "Configuring the Load Process" on page 343
• "Configuring the Match Process" on page 363
• "Configuring the Consolidate Process" on page 443
• "Configuring the Publish Process" on page 449
- 217 -
Chapter 9: Informatica MDM Hub
Processes
Chapter Contents
• "About Informatica MDM Hub Processes" on page 218
• "Land Process" on page 221
• "Stage Process" on page 224
• "Load Process" on page 227
• "Tokenize Process" on page 240
• "Match Process" on page 245
• "Consolidate Process" on page 255
• "Publish Process" on page 260
- 218 -
Note: The publish process is not shown in this figure because it is not a batch
process.
Consolidation Indicator
- 219 -
Value State Name Description
process.
4 NEWLY_ This record is new (load insert) or changed (load
LOADED update) and needs to undergo the match process.
9 ON_HOLD The data steward has put this record on hold until
further notice. Any record can be put on hold regardless
of its consolidation indicator value. The match and
consolidate processes ignore on-hold records. For more
information, see the .
Informatica MDM Hub updates the consolidation indicator for base object
records in the following sequence.
1. During the load process, when a new or updated record is loaded into a
base object, Informatica MDM Hub assigns the record a consolidation
indicator of 4, indicating that the record needs to be matched.
2. Near the start of the match process, when a record is selected as a match
candidate, the match process changes its consolidation indicator to 3.
Note: Any change to the match or merge configuration settings will trigger
a reset match dialog, asking whether you want to reset the records in the
base object (change the consolidation indicator to 4, ready for match). For
more information, see "Configuring the Match Process" on page 363 and
"Configuring the Consolidate Process" on page 443.
3. Before completing, the match process changes the consolidation indicator
of match candidate records to 2 (ready for consolidation).
Note: The match process may or may not have found matches for the
record.
A record with a consolidation indicator of 2 or 4 is visible in Merge
Manager. For more information, see the Informatica MDM Hub Data
Steward Guide.
4. If Accept All Unmatched Rows as Unique is enabled, and a record has
undergone the match process but no matches were found, then
Informatica MDM Hub automatically changes its consolidation indicator to
1 (unique). For more information, see "Accept All Unmatched Rows as
Unique" on page 369.
5. If Accept All Unmatched Rows as Unique is enabled, after the record has
undergone the consolidate process, and once a record has no more
duplicates to merge with, Informatica MDM Hub changes its consolidation
indicator to 1, meaning that this record is unique in the base object, and
that it represents the master record (best version of the truth) for that
entity in the base object.
Note: Once a record has its consolidation indicator set to 1, Informatica
MDM Hub will never directly match it against any other record. New or
- 220 -
updated records (with a consolidation indicator of 4) can be matched
against consolidated records.
Survivorship applies to both trust-enabled columns and columns that are not
trust enabled. When comparing cells from two different records, Informatica
MDM Hub determines survivorship based on the following factors, in order of
precedence:
1. By trust score (only if the two columns are trust-enabled). The data with
the highest trust score wins. If the trust scores are equal, or if trust is not
enabled for both columns, then proceed to the next comparison.
2. By SRC_LUD in the cross-reference record. The data with the more recent
cross-reference SRC_LUD value wins. If the SRC_LUD values are equal,
then proceed to the next comparison.
3. If both records are incoming load updates, then by LAST_UPDATE_DATE
values in the associated cross-reference records. The data with the more
recent cross-reference LAST_UPDATE_DATE wins. If the cross-reference
LAST_UPDATE_DATE values are equal, or if both records are not load
updates, then proceed to the next comparison.
4. By LAST_UPDATE_DATE in the cross-reference record. The data with the
more recent cross-reference LAST_UPDATE_DATE value wins. If the cross-
reference LAST_UPDATE_DATE values are equal, then proceed to the next
comparison.
5. By ROWID_OBJECT in the base object. ROWID_OBJECT values are
evaluated in numeric descending order. The data with the highest ROWID_
OBJECT wins.
Land Process
This section describes concepts and tasks associated with the land process in
Informatica MDM Hub.
- 221 -
About the Land Process
Landing data is the initial step for loading data into Informatica MDM Hub.
Landing data involves the transfer of data from one or more source systems to
Informatica MDM Hub landing tables.
The following figure shows the land process in relation to other Informatica
MDM Hub processes.
- 222 -
Land Process is External to Informatica MDM Hub
The land process is external to Informatica MDM Hub and is executed using an
external batch process or an external application that directly populates
landing tables in the Hub Store. Subsequent processes for managing data are
internal to Informatica MDM Hub.
For any given source system, the approach used depends on whether it is the
most efficient—or perhaps the only—way to data from a particular source
system. In addition, batch processing is often used for the initial data load
(the first time that business data is loaded into the Hub Store), as it can be the
most efficient way to populate the landing table with a large number of
records. For more information, see "Initial Data Loads and Incremental Loads"
on page 229.
- 223 -
Note: Data in the landing tables cannot be deleted until after the load process
for the base object has been executed and completed successfully.
Stage Process
This section describes concepts and tasks associated with the stage process in
Informatica MDM Hub.
- 224 -
data cleansing, if any, to perform on the data before it is saved in the target
table.
The stage process prepares data for the load process, described in "Load
Process" on page 227, which subsequently loads data from the staging table
into a target base object.
The following figure shows the stage process in relation to other Informatica
MDM Hub processes.
The following tables in the Hub Store are associated with the stage process.
Type Description
of
Table
landing Contains data that is copied from a source system. For more
table information, see "About the Land Process" on page 222 and "About
Landing Tables" on page 269.
- 225 -
Type Description
of
Table
staging Contains data that was accepted and copied from the landing table
table during the stage process. For more information, see "About Staging
Tables" on page 275.
raw Contains data that was archived from landing tables. Raw data can be
table configured to get archived based on the number of loads or the
duration (specific time interval). For more information, see
"Configuring the Audit Trail for a Staging Table" on page 300 and
"Configuring Delta Detection for a Staging Table" on page 302.
reject Contains records that Informatica MDM Hub has rejected for a specific
table reason. Records in these tables will not be loaded into base objects.
Data gets rejected automatically during Stage jobs for the following
reasons:
• future date or NULL date in the LAST_UPDATE_DATE column
• NULL value mapped to the PKEY_SRC_OBJECT of the staging table
• duplicates found in PKEY_SRC_OBJECT
- 226 -
Load Process
This section describes concepts and tasks associated with the load process in
Informatica MDM Hub. For related tasks, see "Managing the Load Process" on
page 239.
The load process determines what to do with the data in the staging table
based on:
• whether a corresponding record already exists in the target table and, if
so, whether the record in the staging table has been updated since the load
process was last run
• whether trust is enabled for certain columns (base objects only); if so, the
load process calculates trust scores for the cell data
• whether the data is valid to load; if not, the load process rejects the record
instead
• other configuration settings
- 227 -
Tables Associated with the Load Process
In addition to base objects, the following tables in the Hub Store are
associated with the load process.
Type of Description
Table
staging Contains the data that was accepted and copied from the landing
table table during the stage process. For more information, see "Stage
Process" on page 224 and "About Staging Tables" on page 275.
cross- Used for tracking the lineage of data—the source system for each
reference record in the base object. For each source system record that is
table loaded into the base object, Informatica MDM Hub maintains a
record in the cross-reference table that includes:
• an identifier for the system that provided the record
• the primary key value of that record in the source system
• the most recent cell values provided by that system
Each base object record will have one or more cross-reference
records. For more information, see "Cross-Reference Tables" on
page 86.
history If history is enabled for the base object, and records are updated or
tables inserted, then the load process writes to this information into two
tables:
• base object history table
• cross-reference history table
For more information, see "History Tables" on page 88.
reject Contains records from the staging table that the load process has
table rejected for a specific reason. Rejected records will not be loaded
into base objects. The reject table is associated with the staging
table (called stagingTableName_REJ). For more information, see
"Rejected Records in Load Jobs" on page 238. Rejected records can
be inspected after running Load jobs (see "Viewing Rejected
Records" on page 510).
- 228 -
Initial Data Loads and Incremental Loads
The initial data load (IDL) is the very first time that data is loaded into a
newly-created, empty base object.
During the initial data load, all records in the staging table are inserted into
the base object as new records. For more information, see "Load Inserts" on
page 232.
Once the initial data load has occurred for a base object, any subsequent load
processes are called incremental loads because only new or updated data is
loaded into the base object.
Trust Settings
If a column in a base object derives its data from multiple source systems,
Informatica MDM Hub uses trust to help with comparing the relative reliability
of column data from different source systems. For example, the Orders
system might be a more reliable source of billing addresses than the Direct
Marketing system.
- 229 -
Trust is enabled and configured at the column level. For example, you can
specify a higher trust level for Customer Name in the Orders system and for
Phone Number in the Billing system.
Data stewards can manually override a calculated trust setting if they have
direct knowledge that a particular value is correct. Data stewards can also
enter a value directly into a record in a base object. For more information, see
the Informatica MDM Hub Data Steward Guide.
- 230 -
Validation Rules
When data meets the criterion specified by the validation rule, then the trust
value for that data is downgraded by the percentage specified in the validation
rule. For example:
Downgrade trust on First_Name by 50% if Length < 3
Downgrade trust on Address Line 1, City, State, Zip and Valid_address_ind
if Valid_address_ind= ‘False’
If the Reserve Minimum Trust flag is enabled (checked) for a column, then the
trust cannot be downgraded below the column’s minimum trust setting.
The load process handles staging table records in batches. For each base
object, the load batch size setting (see "Load Batch Size" on page 91) specifies
the number of records to load per batch cycle (default is 1000000).
During execution of the load process for a base object, Informatica MDM Hub
creates a temporary table (_TLL) for each batch as it cycles through records in
the staging table. For example, suppose the staging table contained 250
records to load, and the load batch size were set to 100. During execution, the
load process would:
• create a TLL table and process the first 100 records
• drop and create the TLL table and process the second 100 records
• drop and create the TLL table and process the remaining 50 records
• drop and create the TLL table and stop executing because the TLL table
contained no records
- 231 -
Determining Whether Records Already Exist
During the load process, Informatica MDM Hub first checks to see whether the
record has the same primary key as an existing record from the same source
system. It compares each record in the staging table with records in the target
table to determine whether it already exists in the target table.
Load Description
Operation
load If a record in the staging table does not already exist in the target
insert table, then Informatica MDM Hub inserts that new record in the
target table.
load If a record in the staging table already exists in the target table,
update then Informatica MDM Hub takes the appropriate action. A load
update occurs if the target base object gets updated with data in a
record from the staging table. The load process updates a record
only if it has changed since the record was last supplied by the
source system.
Load updates are governed by current Informatica MDM Hub
configuration settings and characteristics of the data in each record
in the staging table. For example, if Force Update is enabled (see
"Forcing Updates in Load Jobs" on page 544), the records will be
updated regardless of whether they have already been loaded.
During the load process, load updates are executed first, followed by load
inserts.
Load Inserts
What happens during a load insert depends on the target base object and other
factors.
- 232 -
Load Inserts and Target Base Objects
- 233 -
matched to other records in the base object. For more information, see
"Consolidation Status for Base Object Records" on page 219.
• The load process inserts a record into the cross-reference table associated
with the base object. The load process generates a primary key value for
the cross-reference table, then copies into this new record the generated
key, an identifier for the source system, and the columns in the staging
table (including PKEY_SRC_OBJECT). For more information, see "Cross-
Reference Tables" on page 86.
Note: The base object does not contain the primary key value from the
source system. Instead, the base object’s primary key is the generated
ROWID_OBJECT value. The primary key from the source system (PKEY_
SRC_OBJECT) is stored in the cross-reference table instead.
• If history is enabled for the base object (see "History Tables" on page 88),
then the load process inserts a record into its history and cross-reference
history tables.
• If trust is enabled for one or more columns in the base object, then the
load process also inserts records into control tables that support the trust
algorithms, populating the elements of trust and validation rules for each
trusted cell with the values used for trust calculations. This information
can be used subsequently to calculate trust when needed. For more
information, see "Configuring Trust for Source Systems" on page 344 and
"Control Tables for Trust-Enabled Columns" on page 345.
• If Generate Match Tokens on Load is enabled for a base object (see
"Generate Match Tokens on Load" on page 92), then the tokenize process
is automatically started after the load process completes.
Load Updates
What happens during a load update depends on the target base object and
other factors.
- 234 -
• If the record in the staging table has been updated since the last time
the record was supplied by the source system, then the load process
proceeds with the load update.
• If the record in the staging table is unchanged since the last time the
record was supplied by the source system, then the load process
ignores the record (no action is taken) if the dates are the same and
trust is not enabled, or rejects the record if it is a duplicate.
Administrators can change the default behavior so that the load process
bypasses this LAST_UPDATE_DATE check and forces an update of the
records regardless of whether the records might have already been
loaded. For more information, see "Forcing Updates in Load Jobs" on page
544.
• The load process performs foreign key lookups and substitutes any foreign
key value(s) required to maintain referential integrity. For more
information, see "Performing Lookups Needed to Maintain Referential
Integrity" on page 237.
• If the target base object has trust-enabled columns, then the load process:
• calculates the trust score for each trust-enabled column in the record
to be updated, based on the configured trust settings for this trusted
column (as described in "Configuring Trust for Source Systems" on
page 344)
• applies validation rules, if defined, to downgrade trust scores where
applicable (see "Configuring Validation Rules" on page 353)
The load process updates the target record in the base object according to
the following rules:
• If the trust score for the cell in the staging table record is higher than
the trust score in the corresponding cell in the target base object
record, then the load process updates the cell in the target record.
• If the trust score for the cell in the staging table record is lower than
the trust score in the corresponding cell in the target base object
record, then the load process does not update the cell in the target
record.
• If the trust score for the cell in the staging table record is the same as
the trust score in the corresponding cell in the target base object
record, or if trust is not enabled for the column, then the cell value in
the record with the most recent LAST_UPDATE_DATE wins.
- 235 -
• If the staging table record has a more recent LAST_UPDATE_DATE,
then the corresponding cell in the target base object record is
updated.
• If the target record in the base object has a more recent LAST_
UPDATE_DATE, then the cell is not updated.
For more information, see "Survivorship and Order of Precedence" on
page 221.
• For each updated record in the base object, the load process sets its
DIRTY_IND to 1 so that match keys can be regenerated during the tokenize
process. For more information, see "Base Object Records Flagged for
Tokenization" on page 243.
• Whenever an update happens on a base object record, it retains the
consolidation indicator value. For more information, see "Consolidation
Status for Base Object Records" on page 219.
• Whenever the load process updates a record in the base object, it also
updates the associated record in the cross-reference table ("Cross-
Reference Tables" on page 86), history tables (if history is enabled, see
"History Tables" on page 88), and other control tables as applicable.
- 236 -
Performing Lookups Needed to Maintain Referential Integrity
Undefined Lookups
If a lookup on a child object is not defined (the lookup table and column were
not populated), before you can successfully load data, you must repeat the
stage process for the child object prior to executing the load process. For
more information, see "Stage Jobs" on page 556 and "Load Jobs" on page 542.
When configuring columns for a staging table in the Schema Manager, you can
specify whether to allow NULL foreign keys for target base objects. In the
Schema Manager, the Allow Null Foreign Key check box (see "Properties for
Columns in Staging Tables" on page 279) determines whether NULL foreign
keys are permitted.
• By default, the Allow Null Foreign Key check box is unchecked, which
means that NULL foreign keys are not allowed. The load process:
• accepts records valid lookup values
• rejects records with NULL foreign keys
• rejects records with invalid foreign key values
• If Allow Null Foreign Key is enabled (selected), then the load process:
• accepts records with valid lookup values
• accepts records with NULL foreign keys (and permits load inserts and
load updates for these records)
• rejects records with invalid foreign key values
The load process permits load inserts and load updates for accepted records
only. Rejected records are inserted into the reject table rather than being
loaded into the target table.
- 237 -
Note: During the initial data load only, when the target base object is empty,
the load process allows null foreign keys. For more information, see "Initial
Data Loads and Incremental Loads" on page 229.
During the load process, records in the staging table might be rejected for the
following reasons:
• future date or NULL date in the LAST_UPDATE_DATE column
• NULL value mapped to the PKEY_SRC_OBJECT of the staging table
• duplicates found in PKEY_SRC_OBJECT
• invalid value in the HUB_STATE_IND field (for state-enabled base objects
only)
• invalid or NULL foreign keys, as described in "Allowing Null Foreign Keys"
on page 237
Rejected records will not be loaded into base objects. Rejected records can be
inspected after running Load jobs (see "Viewing Rejected Records" on page
510).
For more information about configuring the behavior delta detection for
duplicates and the retention of records in the REJ and RAW tables for a staging
table, see "Using Audit Trail and Delta Detection" on page 300.
Note: To reject records, the load process requires traceability back to the
landing table. If you are loading a record from a staging table and its
corresponding record in the associated landing table has been deleted, then
the load process does not insert it into the reject table.
If the child table contains generated keys from the parent table, the load
process copies the appropriate primary key value from the parent table into
the child table. For example, suppose you had the following data.
PARENT TABLE:
PARENT_ID FNAME LNAME
101 Joe Smith
102 Jane Smith
- 238 -
ADDRESS CITY STATE FKEY_PARENT
1893 my city CA 101
1893 my city CA 102
The load process has special considerations when processing records for
state-enabled base objects. For more information, see "Rules for Loading
Data" on page 168.
Note: The load process rejects any record from the staging table that has an
invalid value in the HUB_STATE_IND column. For more information, see "Hub
State Indicator" on page 160.
Generating match tokens is required before running the match process. In the
Schema Manager, when configuring a base object, you can specify whether to
generate match tokens immediately after the Load job completes, or to delay
tokenizing data until the Match job runs. The setting of the Generate Match
Tokens on Load check box determines when the tokenize process occurs. For
more information, see "When to Generate Match Tokens" on page 242.
- 239 -
Tokenize Process
This section describes concepts and tasks associated with the tokenize process
in Informatica MDM Hub.
Match tokens are stored in the match key table associated with the base
object. For each record in the base object, the tokenize process generates one
or more records in the match key table.
- 240 -
The match key table has the following system columns.
Column Data Type Description
Name (Size)
ROWID_ CHAR (14) Identifies the record for which this match token was
OBJECT generated.
SSA_ CHAR (8) Match key for this record. Encoded representation of the
KEY value in the fuzzy match key column (such as names,
addresses, or organization names) for the associated base
object record. String consists of fixed-length,
compressed, and encoded values built from a combination
of the words and numbers in a name or address.
SSA_ VARCHAR2 Non-encoded (plain text) string representation of the
DATA (500) concatenation of the match column(s) defined in the
associated base object record (Fuzzy Match Key as well as
all fuzzy-match columns and exact-match columns).
Each record in the match key table contains a match token (the data in both
SSA_KEY and SSA_DATA).
The match keys that are generated depend on your configured match settings
and characteristics of the data in the base object. The following example
shows match keys generated from strings using a fuzzy match / search
strategy:
String in Record Generated Match Keys
BETH O'BRIEN MMU$?/$-
BETH O'BRIEN PCOG$$$$
BETH O'BRIEN VL/IEFLM
LIZ O'BRIEN PCOG$$$$
LIZ O'BRIEN SXOG$$$-
LIZ O'BRIEN VL/IEFLM
In this example, the strings BETH O'BRIEN and LIZ O'BRIEN have the same
match key values (PCOG$$$$). The match process would consider these to be
match candidates while searching for match candidates during the match
process.
The tokenize process applies to fuzzy-match base objects only—it does not
apply to exact-match base objects. For fuzzy-match base objects, the
tokenize process allows Informatica MDM Hub to match rows with a degree of
fuzziness—the match need not be identical—just sufficiently similar to be
- 241 -
considered a match. For more information, see match / search strategy in
"Exact-match and Fuzzy-match Base Objects" on page 247)
Match tokens are maintained independently of the match process. The match
process depends on the match tokens in the match key table being current.
Updating match tokens can occur:
• after the load process (see "Generating Match Tokens (Optional)" on page
239), with any changed records (load inserts or load updates)
• when data is put into the base object using SIF Put or CleansePut requests
(see "Generate Match Tokens on Load" on page 92, as well as the
Informatica MDM Hub Services Integration Framework Guide and the
Informatica MDM Hub Javadoc)
• when you run the Generate Match Tokens job (see "Generate Match Tokens
Jobs" on page 540)
• at the start of a match job, as described in "Regenerating Match Tokens If
Needed" on page 251
- 242 -
Base Object Records Flagged for Tokenization
All base objects have a system column named DIRTY_IND. This dirty indicator
identifies when match keys need to be generated for the base object record.
Match tokens are stored in the match key table.
For each record in the base object whose DIRTY_IND is 1, the tokenize
process generates match tokens, and then resets the DIRTY_IND to 0.
The following figure shows how the DIRTY_IND flag changes during various
batch processes:
- 243 -
Key Types and Key Widths in Fuzzy-Match Base Objects
For fuzzy-match base objects, match keys are generated based on the
following settings:
Property Description
key type Identifies the primary type of information being tokenized (Person_
Name, Organization_Name, or Address_Part1) for this base object.
The match process uses its intelligence about name and address
characteristics to generate match keys and conduct searches.
Available key types depend on the population set being used, as
described in "Population Sets" on page 248. For more information,
see "Key Types" on page 392.
key Determines the thoroughness of the analysis of the fuzzy match key,
width the number of possible match candidates returned, and how much
disk space the keys consume. Available key widths are Limited,
Standard, Extended, and Preferred. For more information, see "Key
Widths" on page 392.
Because match keys must be able to overcome errors, variations, and word
transpositions in the data, Informatica MDM Hub generates multiple match
tokens for each name, address, or organization. The number of keys
generated per base object record varies, depending on your data and the
match key width.
The Match Keys Distribution tab in the Match / Merge Setup Details pane of the
Schema Manager allows you to investigate the distribution of match keys in
the match key table. This tool can assist you with identifying potential hot
spots in your data—high concentrations of match keys that could result in
overmatching—where the match process generates too many matches,
including matches that are not relevant. For more information, see
"Investigating the Distribution of Match Keys" on page 438.
Tokenize Ratio
You can configure the match process to repeat the tokenize process whenever
the percentage of changed records exceeds the specified ratio, which is
configured as an advanced property in the base object. For more information,
see "Complete Tokenize Ratio" on page 90.
- 244 -
Managing the Tokenize Process
To manage the tokenize process, refer to the following topics in this
documentation:
Task Topic(s)
Configuration • "Complete Tokenize Ratio" on page 90
• "Generate Match Tokens on Load" on page 92
• "Generating Match Tokens (Optional)" on page 239
Execution "Using Batch Jobs " on page 496
• "Generate Match Tokens Jobs" on page 540
"Writing Custom Scripts to Execute Batch Jobs " on
page 559
• "Generate Match Token Jobs" on page 573
Application Informatica MDM Hub Services Integration Framework
Development Guide
Match Process
This section describes concepts and tasks associated with the match process
in Informatica MDM Hub.
In Informatica MDM Hub, the match process provides you with two main ways
in which to compare records and determine duplicates:
• Fuzzy matching is the most common means used in Informatica MDM Hub
to match records in base objects. Fuzzy matching looks for sufficient
points of similarity between records and makes probabilistic match
determinations that consider likely variations in data patterns, such as
misspellings, transpositions, the combining or splitting of words,
omissions, truncation, phonetic variations, and so on.
- 245 -
• Exact matching is less commonly-used because it matches records with
identical values in the match column(s). An exact strategy is faster, but an
exact match might miss some matches if the data is imperfect.
The best option to choose depends on the characteristics of the data, your
knowledge of the data, and your particular match and consolidation
requirements. For more information, see "Exact-match and Fuzzy-match Base
Objects" on page 247.
During the match process, Informatica MDM Hub compares records in the
base object for points of similarity. If the match process finds sufficient points
of similarity (identical or similar matches) between two records, indicating
that the two records probably are duplicates of each other, then the match
process:
• populates a match table with ROWID_OBJECT references to matched
record pairs, along with the match rule that identified the match, and
whether the matched records qualify for automatic consolidation
- 246 -
Key Concepts for the Match Process
This section describes key concepts that apply to the match process.
Match Rules
A match rule defines the criteria by which Informatica MDM Hub determines
whether two records in the base object might be duplicates. Informatica MDM
Hub supports two types of match rules:
Type Description
Match Used to match base object records based on the values in columns
column you have defined as match columns, such as last name, first name,
rules address1, and address2. This is the most commonly-used method for
identifying matches. For more information, see "Configuring Match
Columns" on page 387.
Primary Used to match records from two systems that use the same primary
key keys for records. It is uncommon for two different source systems to
match use identical primary keys. However, when this does occur, primary
rules key matches are quick and very accurate. For more information, see
"Configuring Primary Key Match Rules" on page 434.
Both kinds of match rules can be used together for the same base object.
- 247 -
Type of Base Description
Object
fuzzy-match Can have both fuzzy match and exact match columns:
base object • fuzzy match only
• exact match only, or
• some combination of fuzzy and exact match
The type of base object determines the type of match and the type of match
columns you can define. The base object type is determined by the selected
match / search strategy for the base object. For more information, see
"Match/Search Strategy" on page 370.
Population Sets
For base objects with the fuzzy match/search strategy, the match process
uses standard population sets to account for national, regional, and language
differences. The population set affects how the match process handles
tokenization, the match / search strategy, and match purposes. For more
information, see "Fuzzy Population" on page 370.
- 248 -
A population set encapsulates intelligence about name, address, and other
identification information that is typical for a given population. For example,
different countries use different address formats, such as the placement of
street numbers and street names, location of postal codes, and so on.
Similarly, different regions have different distributions for surnames—the
surname “Smith” is quite common in the United States population, for
example, but not so common for other parts of the world.
The match for duplicate data functionality is used to generate matches for
duplicates of all non-system base object columns. These matches are
generated when there are more than a set number of occurrences of complete
duplicates on the base object columns (see "Duplicate Match Threshold" on
page 91). For most data, the optimal value is 2.
Note: The Match for Duplicate Data job is visible in the Batch Viewer if the
threshold is set above 1 and there are no NON_EQUAL match rules defined on
the corresponding base object. For more information, see "Match for Duplicate
Data Jobs" on page 552.
After running the match process and creating build match groups, and before
the running consolidation process, you might see the following records:
• record 2 matches to record 1
• record 3 matches to record 1
• record 4 matches to record 1
- 249 -
In this example, there was no explicit rule that matched record 4 to record 1.
Instead, the match was made indirectly due to the behavior of other matches
(record 1 matched to 2, 2 matched to 3, and 3 matched to 4). An indirect
matching is also known as a transitive match. In the Merge Manager and Data
Manager, you can display the complete match history to expose the details of
transitive matches.
You can configure the maximum number of manual matches to process during
batch jobs. Setting a limit helps prevent data stewards from being
overwhelmed with thousands of manual consolidations to process. Once this
limit is reached, the match process stops running run until the number of
records ready for manual consolidation has been reduced. For more
information, see "Maximum Matches for Manual Consolidation" on page 368
and "Consolidate Process" on page 255.
Informatica MDM Hub provides a way to match new data with an existing base
object without actually loading the data into the base object. Rather than run
an entire Match job, you can run the External Match job instead to test for
matches and inspect the results. External Match jobs can process both fuzzy-
match and exact-match rules, and can be used with fuzzy-match and exact-
match base objects. For more information, see "External Match Jobs" on page
535 and "Exact-match and Fuzzy-match Base Objects" on page 247.
For your Informatica MDM Hub implementation, you can increase the
throughput of the match process by running multiple Cleanse Match Servers in
parallel. For more information, see "Configuring Cleanse Match Servers" on
page 308 and the material about distributed Cleanse Match Servers in the
Informatica MDM Hub Installation Guide.
When running very large Match jobs with large match batch sizes, if there is a
failure of the application server or the database, you must re-run the entire
batch. Match batches are a unit. There are no incremental checkpoints. To
address this, if you think there might be a database or application server
failure, set your match batch sizes smaller to reduce the amount of time that
will be spent re-running your match batches. For more information, see
"Number of Rows per Match Job Batch Cycle" on page 368 and "Match Jobs" on
page 547.
- 250 -
Run-Time Execution Flow of the Match Process
This section describes the overall sequence of activities that occur during the
execution of match process. The following figure provides an overview of the
flow, which is determined by the configured match/search strategy for the
base object:
The Merge job executes the match process for a single match batch (see
"Flagging the Match Batch" on page 251). The Auto Match and Merge job cycles
repeatedly until there are no more records to match (no more base object
records with a CONSOLIDATION_IND = 4).
The following base object records are ignored during the match process:
• Records with a CONSOLIDATION_IND of 9 (on hold).
• Records with a PENDING or DELETED status. PENDING records can be
included if explicitly enabled according to the instructions in "Enabling
Match on Pending Records" on page 163.
• Records that are manually excluded according to the instructions in
"Excluding Records from the Match Process" on page 441
When the match process (such as a Match or Auto Match and Merge job)
executes, it first checks to determine whether match tokens need to be
generated for any records in the base object and, if so, generates the match
tokens and updates the match key table. Match tokens will be generated if the
c_repos_table.STRIP_INCOMPLETE_IND flag for the base object is 1, or if any
base object records have a DIRTY_IND=1 (see "Base Object Records Flagged
for Tokenization" on page 243). For more information, see "Match Tokens and
Match Keys" on page 240.
The match process cycles through a series of batches until there are no more
base object records to process. It matches a subset of base object records
(the match batch) against all the records available for matching in the base
object (the match pool). The size of the match batch is determined by the
Number of Rows per Match Job Batch Cycle setting ("Number of Rows per
Match Job Batch Cycle" on page 368).
- 251 -
For the match batch, the match process retrieves, in no specific order, base
object records that meet the following conditions:
• the record has a CONSOLIDATION_IND value of 4 (ready for match)
The load process sets the CONSOLIDATION_IND to 4 for any record that is
new (load insert) or updated (load update).
• the record qualifies based on rule set filtering, if configured (see "Enable
Filtering" on page 403 and "Filtering SQL" on page 403)
In this step, the match process applies the configured match rules to the
match candidates. The match process executes the match rules one at a time,
in the configured order. The match process executes exact-match rules and
exact match-column rules first, then it executes fuzzy-match rules.
The match process continues executing the match rules until there is a match
or there are no more rules to execute.
When all of the records in the match batch have been processed, the match
process adds all of the matches for that group to the match table and changes
CONSOLIDATION_IND=2 for the records in the match batch.
Match Pairs
The match process populates a match table for that base object. Each row in
the match table represents a pair of matched records in the base object. The
match table stores the ROWID_OBJECT values for each pair of matched
- 252 -
records, as well as the identifier for the match rule that resulted in the match,
an automerge indicator, and other information.
- 253 -
Column Data Type Description
Name (Size)
records. PENDING records are to be left as
individual matches.
The Automerge and Autolink jobs processes any
records with an AUTOMERGE_IND of 1. For more
information, see "Automerge Jobs" on page 534 and
"Autolink Jobs" on page 532.
CREATOR VARCHAR2 User or process responsible for creating the record.
(50)
CREATE_ DATE Date on which the record was created.
DATE
UPDATED_BY VARCHAR2 User or process responsible for the most recent
(50) update to the record.
LAST_ DATE Date on which the record was last updated.
UPDATE_
DATE
- 254 -
Task Topic(s)
• "Configuring Primary Key Match Rules" on page 434
• "Investigating the Distribution of Match Keys" on
page 438
• "Excluding Records from the Match Process" on page
441
"Configuring International Data Support" on page 698
• "Configuring Match Settings for Non-US Populations"
on page 699
Execution "Using Batch Jobs " on page 496
• "Auto Match and Merge Jobs" on page 532
• "External Match Jobs" on page 535
• "Generate Match Tokens Jobs" on page 540
• "Key Match Jobs" on page 541
• "Match Jobs" on page 547
• "Match Analyze Jobs" on page 550
• "Match for Duplicate Data Jobs" on page 552
• "Reset Links Jobs" on page 555
• "Reset Match Table Jobs" on page 555
"Writing Custom Scripts to Execute Batch Jobs " on page
559
• "Auto Match and Merge Jobs" on page 570
• "External Match Jobs" on page 572
• "Generate Match Token Jobs" on page 573
• "Key Match Jobs" on page 579
• "Match Jobs" on page 586
• "Match Analyze Jobs" on page 588
• "Match for Duplicate Data Jobs" on page 589
Application Informatica MDM Hub Services Integration Framework
Development Guide
Consolidate Process
This section describes concepts and tasks associated with the consolidate
process in Informatica MDM Hub.
- 255 -
The following figure shows cell data in records from three different source
systems being consolidated into a single master record.
- 256 -
for matches, or the maximum number of records for manual consolidation
is reached.
Traceability
The goal in Informatica MDM Hub is to identify and eliminate all duplicate data
and to merge or link them together into a single, consolidated record while
maintaining full traceability. Traceability is Informatica MDM Hub functionality
that maintains knowledge about which systems—and which records from
those systems—contributed to consolidated records. Informatica MDM Hub
maintains traceability using cross-reference and history tables.
- 257 -
Option Description
for child to specify what happens if records in the parent base object are
base unmerged. For more information, see "Unmerge Child When
objects Parent Unmerges (Cascade Unmerge)" on page 446.
child base For two base objects in a parent-child relationship, if enabled on
object the child base object, child records are resubmitted for the match
records on process if parent records are consolidated. For more information,
parent see "Requeue On Parent Merge" on page 91.
merge
Consolidation Options
There are two ways to consolidate matched records:
• Merging (physical consolidation) combines the matched records and
updates the base object. Merging occurs for merge-style base objects (link
is not enabled).
• Linking (virtual consolidation) creates a logical link between the matched
records. Linking occurs for link-style base objects (link is enabled).
Match rules are configured to identify definite matches for automerging and
close matches for manual merging.
- 258 -
Best Version of the Truth
For a base object, the best version of the truth (sometimes abbreviated as
BVT) is a record that has been consolidated with the best cells of data from the
source records. The precise definition depends on the base object style:
• For merge-style base objects, the base object record is the BVT record,
and is built by consolidating with the most-trustworthy cell values from the
corresponding source records.
• For link-style base objects, the BVT Snapshot job will build the BVT
record(s) by consolidating with the most-trustworthy cell values from the
corresponding linked base object records and return to the requestor a
snapshot for consumption.
- 259 -
Task Topic(s)
• "Manual Unlink Jobs" on page 546
• "Manual Unmerge Jobs" on page 546
• "Multi Merge Jobs" on page 552
• "Reset Links Jobs" on page 555
• "Reset Match Table Jobs" on page 555
• "Synchronize Jobs" on page 557
"Writing Custom Scripts to Execute Batch Jobs " on
page 559
• "Auto Match and Merge Jobs" on page 570
• "Autolink Jobs" on page 570
• "Automerge Jobs" on page 571
• "BVT Snapshot Jobs" on page 572
• "Manual Link Jobs" on page 581
• "Manual Unlink Jobs" on page 582
• "Manual Unmerge Jobs" on page 583
Application Informatica MDM Hub Services Integration Framework
Development Guide
Publish Process
This section describes concepts and tasks associated with the publish process
in Informatica MDM Hub.
- 260 -
Informatica MDM Hub supports two JMS models:
• point-to-point—specific destination for a target external system
• publish/subscribe: point-to-point to an Enterprise Service Bus (ESB),
then publish/subscribe from the ESB to other systems.
In contrast, the publish process belongs to the main Informatica MDM Hub
outbound flow—distribution. Once the master record is established or updated
for a given entity, Informatica MDM Hub can then (optionally) distribute the
master record data to other applications or databases. For an introduction to
reconciliation and distribution, see the Informatica MDM Hub Overview. In
another scenario, data changes can be sent to the Activity Manager Rules
queue so that the data change can be evaluated against user-defined rules.
The land, stage, load, match, and consolidate processes work with batches of
records and are executed as batch jobs or stored procedures. In contrast, the
publish process is executed as the result of a message trigger that executes
when a data change occurs in the Hub Store. The message trigger creates an
XML message that gets published on a JMS message queue.
- 261 -
those JNDI names that have been set up by the application server. The Hub
Console allows you to register message queue servers and message queues
that have already been configured in the application server environment.
In this scenario:
1. A batch load or a real-time SIF API request (SIF put or cleanse_put
request) may result in an insert or update on a base object.
- 262 -
You can configure a message rule to control data going to the C_REPOS_
MQ_DATA_CHANGE table.
2. Hub Server polls data from C_REPOS_MQ_DATA_CHANGE table at regular
intervals.
3. For data that has not been sent, Hub Server constructs an XML message
based on the data and sends it to the outbound queue configured for the
message queue.
4. It is the external application's responsibility to retrieve the message from
the outbound queue and process it.
- 263 -
Chapter 10: Configuring the Land
Process
This chapter explains how to configure the land process for your Informatica
MDM Hub implementation. For an introduction, see "Land Process" on page
221.
Chapter Contents
• "Before You Begin" on page 264
• "Configuration Tasks for the Land Process" on page 264
• "Configuring Source Systems" on page 264
• "Configuring Landing Tables" on page 269
- 264 -
About Source Systems
Source systems are external applications or systems that provide data to
Informatica MDM Hub. In order to manage input from various source systems,
Informatica MDM Hub requires a unique internal name for each source
system. You use the Systems and Trust tool in the Model workbench to define
source systems for your Informatica MDM Hub implementation.
If multiple source systems contribute data for the same column in a base
object, you can configure trust on a column-by-column basis to specify which
source system(s) are more reliable providers of data (relative to other source
systems) for that column. Trust is used to determine survivorship when two
records are consolidated, and whether updates from a source system are
sufficiently reliable to update the “best version of the truth” record. For more
information, see "Configuring Trust for Source Systems" on page 344.
Informatica MDM Hub uses an administration source system for manual trust
overrides and data edits from the Data Manager or Merge Manager tools,
which are described in the Informatica MDM Hub Data Steward Guide. This
administration source system can contribute data to any trust-enabled
column. The administration source system is named Admin by default, but you
can optionally change its name according to the instructions in "Editing Source
System Properties" on page 267.
The source systems that you define in the Systems and Trust tool are stored in
a special public Informatica MDM Hub repository table (C_REPOS_SYSTEM,
with a display name of MRM System). This table is visible in the Schema
Manager if the Show System Tables option is selected (for more information,
see "Changing the Item View" on page 40). C_REPOS_SYSTEM can also be
used in packages, as described in "Configuring Packages" on page 151.
- 265 -
• In the Hub Console, expand the Model workbench, and then click Systems
and Trust.
Property Description
Name Unique, descriptive name for this source system.
Primary Primary key for this source system. Unique identifier for this
Key system in the ROWID_SYSTEM column of C_REPOS_SYSTEM. Read
only.
Description Optional description for this source system.
- 266 -
4. Specify the source system properties. For more information, see "Source
System Properties" on page 266.
5. Click OK.
The Systems and Trust tool displays the newly-added source system in the
list of source systems.
Note: When you add a source system, Hub Store uses the first 14
characters of the system name (in all uppercase letters) as its primary key
(ROWID_SYSTEM value in C_REPOS_SYSTEM).
Note: If this source system has already contributed data to your Informatica
MDM Hub implementation, Informatica MDM Hub continues to track the
lineage (history) of data from this source system even after you have
renamed it.
The screen refreshes, showing the Edit button next to the name and
description fields for the selected source system.
- 267 -
4. Change any of the editable properties. For more information, see "Source
System Properties" on page 266.
5. To change trust settings for a source system, see "Configuring Trust for
Source Systems" on page 344.
Note: Removing a source system deletes only the source system definition in
the Hub Console—it has no effect outside of Informatica MDM Hub.
- 268 -
The Systems and Trust tool removes the source system from the list,
along with any metadata associated with this source system.
The manner in which source systems populate landing tables with data is
entirely external to Informatica MDM Hub. The data model you use for
collecting data in landing tables from various source systems is also external
to Informatica MDM Hub. One source system could populate multiple landing
tables. A single landing table could receive data from different source
systems. The data model you use is entirely up to your particular
implementation requirements.
Inside Informatica MDM Hub, however, landing tables are mapped to staging
tables, as described in "Mapping Columns Between Landing and Staging
Tables" on page 286. It is in the staging table—mapped to a landing table—
where the source system supplying the data to the base object is identified.
During the load process, Informatica MDM Hub copies data from a landing
table to a target staging table, tags the data with the source system
identification, and optionally cleanses data in the process. A landing table can
be mapped to one or more staging tables. A staging table is mapped to only
one landing table.
- 269 -
Column Description
Type
system Columns that are automatically created and maintained by the
columns Schema Manager.
user- Columns that have been added by users according to the
defined instructions in "Configuring Columns in Tables" on page 102.
columns
Property Description
Item Type Type of table that you are adding. Select Landing Table.
Display Name of this landing table as it will be displayed in the Hub
Name Console.
Physical Actual name of the landing table in the database. Informatica
Name MDM Hub will suggest a physical name for the landing table based
on the display name that you enter.
Data Name of the data tablespace for this landing table. For more
Tablespace information, see the Informatica MDM Hub Installation Guide.
Index Name of the index tablespace for this landing table. For more
Tablespace information, see the Informatica MDM Hub Installation Guide.
Description Description of this landing table.
Create Date and time when this landing table was created.
Date
Contains Specifies whether this landing table contains the full data set from
Full the source system, or only updates.
Data Set • If selected (default), indicates that this landing table contains
the full set of data from the source system (such as for the
initial data load). When this check box is enabled, you can
configure Informatica MDM Hub’s delta detection feature (see
"Configuring Delta Detection for a Staging Table" on page 302)
so that, during the stage process, only changed records are
copied to the staging table.
• If not selected, indicates that this landing table contains only
changed data from the source system (such as for incremental
loads). In this case, Informatica MDM Hub assumes that you
- 270 -
Property Description
filtered out unchanged records before populating the landing
table. Therefore, the stage process inserts all records from
the landing table directly into the staging table. When this
check box is enabled, Informatica MDM Hub’s delta detection
feature is not available.
Note: You can change this property only when editing the source
system properties, as described in "Editing Source System
Properties" on page 267.
- 271 -
5. Specify the properties (described in "Landing Table Properties" on page
270) for this new landing table.
6. Click OK.
The Schema Manager creates the new landing table in the Operational
Reference Store (ORS), along with support tables, and then adds the new
landing table to the schema tree.
7. Configure the columns for your landing table according to the instructions
in "Configuring Columns in Tables" on page 102.
8. If you want to configure this landing table to contain only changed data
from the source system (Contains Full Data Set), edit the landing table
properties according to the instructions in "Editing Landing Table
Properties" on page 272.
- 272 -
4. Change the landing table properties you want. For more information, see
"Landing Table Properties" on page 270.
- 273 -
Chapter 11: Configuring the Stage
Process
This chapter explains how to configure the data staging process for your
Informatica MDM Hub implementation. For an introduction, see "Stage
Process" on page 224. In addition, to learn about cleansing data during the
data staging process, see "Configuring Data Cleansing" on page 307
Chapter Contents
• "Before You Begin" on page 274
• "Configuration Tasks for the Stage Process" on page 274
• "Configuring Staging Tables" on page 275
• "Mapping Columns Between Landing and Staging Tables" on page 286
• "Using Audit Trail and Delta Detection" on page 300
- 274 -
Configuring Staging Tables
This section describes how to configure staging tables in your Informatica
MDM Hub implementation.
The structure of a staging table is directly based on the structure of the target
object that will contain the consolidated data. You use the Schema Manager in
the Model workbench to configure staging tables.
Note: You must have at least one source system defined before you can
define a staging table. For more information, see "Configuring Source
Systems" on page 264.
- 275 -
Physical Data Description
Name Type
(Size)
ROWID_ CHAR Primary key. Unique value assigned by Informatica
OBJECT (14) during the stage process.
DELETED_ INT Reserved for future use.
IND
DELETED_ DATE Reserved for future use.
DATE
DELETED_ VARCHAR Reserved for future use.
BY (50)
LAST_ DATE Date on which the record was last updated in the source
UPDATE_ system. For base objects, this will populate LAST_
DATE UPDATE_DATE and SRC_LUD in the cross-reference
table, and (depending on trust settings) may also
populate LAST_UPDATE_DATE on the base object.
UPDATED_ VARCHAR User or process responsible for the most recent update.
BY (50)
CREATE_ DATE Date on which the record was created.
DATE
CREATOR VARCHAR User or process responsible for creating the record.
(50)
SRC_ VARCHAR Database internal Rowid column that is used to uniquely
ROWID (30) trace back records to the Landing table from Staging.
HUB_ INT For state-enabled base objects only. Integer value
STATE_ indicating the state of this record. Valid values are:
IND • 0=Pending
• 1=Active (Default)
• -1=Deleted
For details, see "Hub State Indicator" on page 160.
Staging tables must be based on the columns provided by the source system
for the target base object for which the staging table is defined, even if the
landing tables are shared across multiple source systems. If you do not make
the column on staging tables source-specific, then you create unnecessary
trust and validation requirements.
If you limit the columns in the staging tables to the columns actually provided
by the source systems, then you can restrict the trust columns to those that
come from two or more staging tables. Use this approach instead of treating
every column as if it comes from every source, which would mean needing to
add trust for every column, and then validation rules to downgrade the trust
on null values for all of the sources that do not provide values for the columns.
More trust columns and validation rules obviously affect the load and the
merge processes. Also, the more trusted columns, the longer will the update
statements be for the control table. Bear in mind that Oracle and DB2 have a
- 276 -
32K limit on the size of the SQL buffer for SQL statements. For this reason,
more than 40 trust columns result in a horizontal split in the update of the
control table—MRM will try to update only 40 columns at a time.
By default, this option is not enabled. During Informatica MDM Hub stage jobs
(see "Stage Jobs" on page 556), for each inbound record of data, Informatica
- 277 -
MDM Hub generates an internal key that it inserts in the ROWID_OBJECT
column of the target base object.
Enable this option when you want to use the value from the primary key
column from the source system instead of Informatica MDM Hub’s internally-
generated key. To enable this option, when adding a staging table to a base
object (see "Adding Staging Tables" on page 280), check (select) the Preserve
Source System Keys check box in the Add staging to Base Object dialog. Once
enabled, during stage jobs, instead of generating an internal key, Informatica
MDM Hub takes the value in the PKEY_SOURCE_OBJECT column from the
staging table and inserts it into the ROWID_OBJECT column in the target base
object.
Note: Once a base object is created, you cannot change this setting.
Note: During the stage process, if multiple records contain the same PKEY_
SRC_OBJECT, the surviving record is the one with the most recent LAST_
UPDATE_DATE. The other records are sent to the reject table. For more
information, see "Tables Associated With the Stage Process" on page 225 and
"Survivorship and Order of Precedence" on page 221.
If the Preserve Source System Keys check box is enabled, then the Schema
Manager displays the Highest Reserved Key field. If you want to insert a gap
between the source key and Informatica MDM Hub’s key, then enter the
amount by which the key is increased after the first load.
Note: Set the Highest Reserved Key to the upper boundary of the source
system keys. To allow a margin, set this number slightly higher, adding a
buffer to the expected range of source system keys. Any records added to the
base object that do not contain this key will be given a key by Informatica
MDM Hub that is above the highest reserved value you set.
Enabling this option has the following consequences when the base object is
first loaded:
1. From the staging table, Informatica MDM Hub takes the value in PKEY_
SOURCE_OBJECT and inserts that into the base object’s ROWID_OBJECT—
instead of generating Informatica MDM Hub’s internal key.
2. Informatica MDM Hub then resets the key's starting position to MAX
(PKEY_SOURCE_OBJECT) + the GAP value.
3. On the next load for this staging table, Informatica MDM Hub continues to
use the PKEY_SOURCE_OBJECT. For loads from other staging tables, it
uses the Informatica MDM Hub-generated key.
- 278 -
Note: Only one staging table per base object can have this option enabled
(even if it is from the same system). The reserved key range is set at the
initial load only.
By default, during the stage process (see "Stage Jobs" on page 556), for each
inbound record of data, Informatica MDM Hub replaces the cell value in the
target base object whenever an incoming record has a higher trust level—
even if the value it replaces is identical. Even though the value has not
changed, Informatica MDM Hub updates the last update date for the cell to the
date associated with the incoming record, and assigns to the cell the same
trust level as a new value. For more information, see "Configuring Trust for
Source Systems" on page 344.
You can change this behavior by checking (selecting) the Cell Update check
box when configuring a staging table. If cell update is enabled, then during
Stage jobs, Informatica MDM Hub will compare the cell value with the current
contents of the cross-reference table before it updates the target record in the
base object. If the cross-reference record for this system has an identical
value in this cell, then Informatica MDM Hub will not update the cell in the Hub
Store. Enabling cell update can increase performance during Stage jobs if your
Informatica MDM Hub implementation does not require updates to the last
update date and trust value in the target base object record.
- 279 -
Property Description
Allow Determines whether null foreign keys are allowed. Use this option
Null only if null values are valid for the foreign key relationship—that is,
Foreign if the foreign key is an optional relationship. For more information,
Key see "Configuring Lookups For Foreign Key Columns" on page 283.
• Check (select) this check box to allow data to be loaded when the
child record does not contain a value for the lookup operation.
• Uncheck (clear, the default) this check box to prevent null
foreign keys. In this case, records with null values in the lookup
column will be written to the rejects table instead of being
loaded.
6. Specify the staging table properties. For more information, see "Staging
Table Properties" on page 277.
- 280 -
Note: Some of these settings cannot be changed after the staging table
has been added, so make sure that you specify the settings you want
before closing this dialog.
7. From the list of the columns in the base object, select all of the columns
that this source system will provide. For more information, see "Staging
Table Columns" on page 275.
• Click the Select All button to select all of the columns without
needing to click each column individually.
row and click the button to define the lookup column. For more
information, see "Configuring Lookups For Foreign Key Columns" on page
283.
Note: You will not be able to save this new staging table unless you
complete this step.
10. Click OK.
The Schema Manager creates the new staging table in the Operational
Reference Store (ORS), along with any support tables, and then adds the
new staging table to the schema tree.
11. If you want, configure an Audit Trail and Delta Detection for this staging
table. For more information, see "Using Audit Trail and Delta Detection" on
page 300.
- 281 -
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. In the schema tree:
4. expand the Base Objects node, and then expand the node for the base
object associated with this staging table.
• If the staging table is associated with the base object, then expand the
Staging Tables node to display it.
5. Select the staging table that you want to configure.
The Schema Manager displays the properties for the selected table.
6. Specify the staging table properties. For more information, see "Staging
Table Properties" on page 277.
For each property that you want to edit (Display Name and Description),
click the Edit button next to it, and specify the new value.
Note: You can change the source system only if the staging table and its
related support tables (raw, opl, and prl tables) are empty.
Should not be able to change the source system if the Staging table (or its
related tables) contain data.
7. From the list of the columns in the base object, change the columns that
this source system will provide.
• Click the Select All button to select all of the columns without
needing to click each column individually.
- 282 -
9. If you want, change lookups for foreign key columns. Select the column
and click the button to configure the lookup column. For more
information, see "Configuring Lookups For Foreign Key Columns" on page
283.
10. If you want to change cell updating (see "Enabling Cell Update" on page
279), click in the Cell update check box.
11. Change the column configuration for your staging table, if you want. For
more information, see "Configuring Columns in Tables" on page 102.
12. If you want, configure an Audit Trail and Delta Detection for this staging
table. For more information, see "Using Audit Trail and Delta Detection" on
page 300.
The Hub Console launches the Systems and Trust tool and displays the source
system associated with this staging table. For more information, see
"Configuring Source Systems" on page 264.
About Lookups
A lookup is the process of retrieving a data value from a parent table during
Load jobs. In Informatica MDM Hub, when configuring a staging table
associated with a base object, if a foreign key column in the staging table (as
the child table) is related to the primary key in a parent table, you can
configure a lookup to retrieve data from that parent table. The target column
in the lookup table must be a unique column (such as the primary key). For
more information, see "Performing Lookups Needed to Maintain Referential
Integrity" on page 237.
For example, suppose your Informatica MDM Hub implementation had two
base objects: a Consumer parent base object and an Address child base
object, with the following relationship between them:
Consumer.Rowid_object = Address.Consumer_Fkey
- 283 -
In this case, the Consumer_Fkey will be included in the Address Staging table
and it will look up data on some column.
Once defined, when the Load job runs on the base object, Informatica MDM
Hub looks up the source system’s Consumer code value in the primary key
from source system column of the Consumer code cross-reference table, and
returns the customer type ROWID_OBJECT value that corresponds to the
source consumer type.
Configuring Lookups
The Edit Lookup button is enabled only for foreign key columns.
- 284 -
The Define Lookup dialog contains the parent base object and its cross-
reference table, along with any unique columns (only).
8. Select the target column for the lookup.
• To define the lookup to a base object, expand the base object and
select Rowid_Object (the primary key for this base object).
- 285 -
13. Click the button to save your changes.
- 286 -
Relationships Between Landing and Staging Tables
You can map columns from one landing table to multiple staging tables.
However, each staging table is mapped to only one landing table.
For each column of data in the staging table, the data comes from the landing
column in one of two ways:
Copy Description
Method
passed Informatica MDM Hub copies the data as is, without making any
through changes to it. Data comes directly from a column in the landing
table.
cleansed Informatica MDM Hub standardizes and verifies data using cleanse
functions. The output of the cleanse function becomes the input to
the target column in the staging table. For more information about
cleanse functions, see "Configuring Data Cleansing" on page 307
In the following figure, data in the Name column is cleansed via a cleanse
function, while data from all other columns is passed directly to the
corresponding target column in the staging table.
Note: A staging table does not need to use every column in the landing table
or every output string from a cleanse function. The same landing table can
provide input to multiple staging tables, and the same cleanse function can be
reused for multiple columns in multiple landing tables.
Cleanse functions can also decompose and aggregate data. Either way, your
mappings need to accommodate the required inputs and outputs.
In the following figure, the cleanse function decomposes the name field,
breaking the data into smaller pieces.
This cleanse function has one input string and five output strings. In your
mapping, you need to make sure that the input string is mapped to the cleanse
function, and each output string is mapped to the correct target column in the
staging table.
- 287 -
Cleanse Functions that Aggregate Data
In the following figure, the cleanse function aggregates data from five fields
into a single string.
This cleanse function has five input strings and one output string. In your
mapping, you need to make sure that the input strings are mapped to the
cleanse function and the output string is mapped to the correct target column
in the staging table.
When you select a mapping in the mappings list, its properties are displayed.
- 288 -
Tabs in the Mappings Tool
When a mapping is selected, the Mappings tool displays the following tabs.
Column Description
General General properties for this mapping. For more information, see
"Mapping Properties" on page 289.
Diagram Interactive diagram that lets you define mappings between
columns in the landing and staging tables. For more information,
see "Mapping Columns Between Landing and Staging Table
Columns" on page 291.
Query Allows you to specify query parameters for this mapping. For
Parameters more information, see "Configuring Query Parameters for
Mappings" on page 294.
Test Allows you to test the mapping.
Mapping Diagrams
When you click the Diagram tab for a mapping, the Mappings tool displays the
current column mappings.
Mapping lines show the mapping from source columns in the landing table to
target columns in the staging table. Colors in the circles at either end of the
mapping lines indicate data types.
Mapping Properties
Mappings have the following properties.
Field Description
Name Name of this mapping as it will be displayed in the Hub Console.
Description Description of this mapping.
Landing Select the landing table that will be the source of the mapping.
Table
Staging Select the staging table that will be the target of the mapping.
Table
Secure Check (enable) to make this mapping a secure resource, which
Resource allows you to control access to this mapping. Once a mapping is
designated as a secure resource, you can assign privileges to it in
the Secure Resources tool. For more information, see "Securing
Informatica MDM Hub Resources" on page 629, and "Assigning
- 289 -
Field Description
Resource Privileges to Roles" on page 641.
Adding Mappings
To create a new mapping:
1. Start the Mappings tool according to the instructions in "Starting the
Mappings Tool" on page 288.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Right-click in the area where the mappings are listed and choose Add
Mapping.
The Mappings tool displays the Mapping dialog.
columns with the same name in the staging table, click the button.
7. Click OK.
8. When you are finished, click the button to save your changes.
Copying Mappings
To create a new mapping by copying an existing one:
- 290 -
1. Start the Mappings tool according to the instructions in "Starting the
Mappings Tool" on page 288.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Right-click the mapping that you want to copy, and then choose Copy
Mapping.
The Mappings tool displays the Mapping dialog.
4. Specify the mapping properties. The landing table is already specified. For
more information, see "Mapping Properties" on page 289.
5. Click OK.
- 291 -
For each mapping:
• inputs are columns from the landing table
• outputs are the columns in the staging table
The workspace and the methods of creating a mapping are the same as for
creating cleanse functions. To learn how to use the workspace to define
functions, inputs, and outputs, see "Configuring Graph Functions" on page 321.
Note: If you want to load by rowid, create a mapping between the primary
key in the landing table and the Rowid object in the staging table. For more
information, see "Loading by RowID" on page 296.
- 292 -
3. Click the button to save your changes.
To cleanse data during Stage jobs, you can include one or more cleanse
functions in your mapping. This section provides brief instructions for
configuring cleanse functions in mappings. For more information, see "Using
Cleanse Functions" on page 314.
- 293 -
5. Click the button to save your changes.
Note: For column mappings (from landing to staging tables) that use cleanse
functions, cleanse functions can be automatically removed from the mappings
in the following circumstances:
• If you change cleanse engines in your Informatica MDM Hub
implementation and column mappings use cleanse functions that are not
available in the new cleanse engine. Unsupported cleanse functions are
automatically removed.
• If you restart the application server for the Cleanse Match Server and the
cleanse engine fails to initialize for some reason. Even after you resolve
the issue(s) that cause the cleanse engine initialization failure, unavailable
cleanse functions are automatically removed.
In either case, you will need to use the Mappings tool in the Hub Console to
reconfigure the mappings using cleanse functions that are supported in the
current cleanse engine.
- 294 -
5. If you want, check or uncheck the Enable Distinct check box, as
appropriate, to configure distinct mapping. For more information, see
"Distinct Mapping" on page 295.
6. If you want, check or uncheck the Enable Condition check box, as
appropriate, to configure conditional mapping. For more information, see
"Conditional Mapping" on page 296.
If enabled, type the SQL WHERE clause (omitting the WHERE keyword),
and then click Validate to validate the clause.
By default, all records are retrieved from the landing table. Optionally, you
can configure a mapping that filters records in the landing table. There are two
types of filters: distinct and conditional. You configure these settings on the
Query Parameters tab in the Mappings tool. For more information, see
"Configuring Query Parameters for Mappings" on page 294.
Distinct Mapping
If you click the Enable Distinct check box on the Query Parameters tab, the
Stage job selects only the distinct records from the landing table. Informatica
MDM Hub populates the staging table using the following SELECT statement:
- 295 -
mapping prevents the two extra customer records from being written to the
rejects table.
In the mapping to the customer table, check (select) Enable Distinct to avoid
having duplicate records because only LUD, CUST_ID, and NAME are mapped
to the Customer staging table. With Distinct enabled, only one record would
populate your customer table and no rejects would occur.
Alternatively, for the address mapping, you map ADDR_ID and ADDR with
Distinct disabled so that you get two records and no rejects.
Conditional Mapping
If you select the Enable Condition check box, you can apply a SQL WHERE
clause to unload the data in cleanse. For example, suppose the data in your
landing table is from all states in the US. You can use the WHERE clause to
filter the data that is written to the staging tables to include only data from
one state, such as California. To do this, type in a WHERE clause (but omit the
WHERE keyword): STATE = 'CA'. When the cleanse job is run, it unloads and
processes records as SELECT * FROM LANDING WHERE STATE = 'CA'. If you
specify conditional mapping, click the Validate button to validate the SQL
statement.
Loading by RowID
You can streamline load, match, and merge processing by explicitly
configuring Informatica MDM Hub to load by RowID. Otherwise, Informatica
MDM Hub loads data according to its default behavior, which is described in
"Run-time Execution Flow of the Load Process" on page 231.
Note: If you clean the BASE OBJECT using the stored procedure, and if you
had setup the TAKE-ON GAP for the particular staging table, the ROWID
sequences are reset to 1.
In the staging table, the Rowid Object column (a nullable column) has a
specialized usage. You can streamline load, match, and merge processing by
mapping any column in a landing table to the Rowid Object column in a staging
table. In the following example, the Customer Id column in the landing table is
mapped to the Rowid Object column in the staging table.
- 296 -
Mapping to the Rowid Object column allows for the loading of records by
present- or lineage-based ROWID_OBJECT. During the load, if an incoming
record with a populated ROWID_OBJECT is new (the incoming PKEY_SRC_
OBJECT + ROWID_SYSTEM is checked), then this record bypasses the match
and merge process and gets added to the base object directly—a real-time API
PUT(_XREF) by ROWID_OBJECT. Using this feature enhances lineage and
unmerge support, enables closed-loop integration with downstream systems,
and can increase throughput.
The initial data load for a base object inserts all records into the target base
object. Therefore, enable loading by rowID for incremental loads that occur
after the initial data load. For more information, see "Initial Data Loads and
Incremental Loads" on page 229 and "Run-time Execution Flow of the Load
Process" on page 231.
Jumping to a Schema
The Mappings tool allows you to quickly launch the Schema Manager and
display the schema associated with the selected mapping.
- 297 -
• by Mapping
4. Right-click anywhere in the navigation pane, and then choose Jump to
Schema.
The Mappings tool displays the schema for the selected mapping.
Testing Mappings
To test a mapping that you have configured:
1. Start the Mappings tool according to the instructions in "Starting the
Mappings Tool" on page 288.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Select the mapping that you want to configure.
4. Click the Test tab.
The Mappings tool displays the Test tab for this mapping.
- 298 -
5. Specify input values for the columns under Input Name.
6. Click Test.
7. The Mappings tool tests the mapping and populates the columns under
Output Name with the results.
Removing Mappings
To remove a mapping:
1. Start the Mappings tool according to the instructions in "Starting the
Mappings Tool" on page 288.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Right-click the mapping that you want to remove, and choose Delete
Mapping.
The Mappings tool prompts you to confirm deletion.
4. Click Yes.
The Mappings tool drops supporting tables, removes the mapping from the
metadata, and updates the list of mappings.
- 299 -
Using Audit Trail and Delta Detection
After you have completed mapping columns between landing and staging
tables, you can configure the audit trail and delta detection features for a
staging table. For more information, see "Mapping Columns Between Landing
and Staging Tables" on page 286.
To configure audit trail and delta detection, click the Settings tab.
Note: The Audit Trail has very different functionality from—and is not to be
confused with—the Audit Manager tool described in "Auditing Informatica MDM
Hub Services and Events" on page 684.
- 300 -
5. At the bottom of the properties panel, click Preserve an audit trail in
the raw table to enable the raw data audit trail.
The Schema Manager prompts you to select the retention period for the
audit table.
Option Description
Loads Number of batch loads for which to retain data.
Time Period Period of time for which to retain data.
Once configured, the audit trail keeps data for the retention period that you
specified. For example, suppose you configured the audit trail for two loads
(Stage job executions). In this case, the audit trail will retain data for the two
most recent loads to the staging table. If there were ten records in each load
in the landing table, then the total number of records in the RAW table would
be 20.
If the Stage job is run multiple times, then the data in the RAW table will be
retained for the most recent two sets based on the ROWID_JOB. Data for older
ROWID_JOBs will be deleted. For example, suppose the value of the ROWID_
JOB for the first Stage job is 1, for the second Stage job is 2, and so on. When
you run the Stage job a third time, then the records in which ROWID_JOB=1
will be discarded.
Note: Using the Clear History button in the Batch Viewer after the first run of
the process:
If the audit trail is enabled for a staging table and you choose the Clear
History button in the Batch Viewer while the associated stage job is selected,
the records in the RAW and REJ tables will be cleared the next time the stage
job is run.
- 301 -
Configuring Delta Detection for a Staging Table
If you enable delta detection for a staging table, Informatica MDM Hub
processes only new or changed records and ignores unchanged records.
- 302 -
6. A list of available columns is displayed. Choose which ones you want to
use for delta detection.
When loading data to stage table, if any column from defined set has the value
different from available previous load value, the row is treated as changed. If
all columns from defined set are the same, the row is treated as unchanged.
Columns that are not mapped are ignored
- 303 -
5. Specify the manner in which you want to have deltas detected.
Option Description
Detect del- All columns are selected for delta comparison, includ-
tas by com- ing the Last Update Date.
paring all
columns in
mapping
Detect del- If your schema has an applicable date column, choose
tas via a this option and select the date column you want to
date column use for delta comparison. This is the preferred option
in cases where you have an applicable date column.
If delta detection is enabled, then the Stage job compares the contents of the
landing table—which is mapped to the selected staging table—against the data
- 304 -
set processed in the previous run of the stage job. This comparison is done to
determine whether the data has changed since the previous run. Changed,
new records, and rejected records will be put into the staging table. Duplicate
records are ignored. For more information, see "Mapping Columns Between
Landing and Staging Tables" on page 286.
Note: Reject records move from cleanse to load after the second stage run.
- 305 -
values that are not the last update date or part of the concatenated
primary key will not be detected.
• Duplicate primary keys are not considered during subsequent stage
processes when using delta detection by mapped columns.
• Reject handling allows you to:
• View all reject records for a given staging table regarding of the batch
job
• View all reject records by day across all staging tables
• Query reject tables based on query filters
- 306 -
Chapter 12: Configuring Data
Cleansing
This chapter describes how to configure your Hub Store to cleanse data during
the stage process. This chapter is a companion to the material provided in
"Configuring the Stage Process" on page 274.
Chapter Contents
• "Before You Begin" on page 307
• "About Data Cleansing in Informatica MDM Hub" on page 307
• "Configuring Cleanse Match Servers" on page 308
• "Using Cleanse Functions" on page 314
• "Configuring Cleanse Lists" on page 333
- 307 -
Note: Data cleansing that occurs prior to its arrival in the landing tables is
outside the scope of this chapter.
The Cleanse Match Server is multi-threaded so that each instance can process
multiple requests concurrently. It can be deployed on a variety of application
servers. See the Informatica MDM Hub Release Notes for a list of supported
application servers. See the Informatica MDM Hub Installation Guide for
instructions on installing and configuring Cleanse Match Server(s).
Informatica MDM Hub supports running multiple Cleanse Match Servers for
each Operational Reference Store (ORS). The cleanse process is generally
CPU-bound. This scalable architecture allows you to scale your Informatica
MDM Hub implementation as the volume of data increases. Deploying Cleanse
Match Servers on multiple hosts distributes the processing load across
multiple CPUs and permits the running of cleanse operations in parallel. In
addition, some external adapters are inherently single-threaded, so this
Informatica MDM Hub architecture allows you to simulate multi-threaded
operations by running one processing thread per application server instance.
- 308 -
• Online and Batch (default)
• Online Only
• Batch Only
For your Informatica MDM Hub implementation, you can increase the
throughput of the cleanse process by running multiple Cleanse Match Servers
in parallel. To learn more about distributed Cleanse Match Servers, see the
Informatica MDM Hub Installation Guide.
If proxy users have been configured for your Informatica MDM Hub
implementation, if you created proxy_user and cmx_ors with different
passwords, then you need to either:
• restart the application server and log in to the proxy user from the Hub
Console
or
• register the Cleanse Match Server for the proxy user again
Cleanse Requests
All requests for cleansing are issued by database stored procedures. These
stored procedures package a cleanse request as an XML payload and transmit
it to a Cleanse Match Server. When the Cleanse Match Server receives a
request, it parses the XML and invokes the appropriate code:
Mode Description
Type
On-line The result is packaged as an XML response and sent back via an
operations HTTP POST connection.
Batch jobs The Cleanse Match Server pulls the data to be processed into a flat
file, processes it, and then uses a bulk loader to write the data
back.
• For Oracle, it uses the Oracle loader (SQLLDR) utility.
• For DB2, it uses the DB2 Load utility.
- 309 -
The Cleanse Match Server is multi-threaded so that each instance can process
multiple requests concurrently. The default timeout for batch requests from
Oracle to a Cleanse Match Server is one year, and the default timeout for on-
line requests is one minute. For DB2, the default timeout for batch requests or
SIF requests is 600 seconds (10 minutes).
When running a stage/match job, if more than one cleanse match server is
registered, and if the total number of records to be staged or matched is more
than 500, then the job will get distributed in parallel among the available
Cleanse Match Servers.
The Cleanse Match Server tool displays a list of any configured Cleanse Match
Servers.
- 310 -
Property Description
Match Determines whether to use the Match Server for matching data.
Server • Check (select) this check box to use the Match Server for
matching data.
• Uncheck (clear) this check box if you do not want to use the
Match Server for matching data.
If an ORS has multiple associated Cleanse Match Servers, you can
enhance performance by configuring each Cleanse Match Server as
either a match-only or a cleanse-only server. Use this option in
conjunction with the Cleanse Server check box to implementation
this configuration.
Match Mode that the Match Server uses for matching data. One of the
Mode following values:
For details, see "Cleanse Requests" on page 309.
Offline Determines whether the Cleanse Match Server is offline or online.
• Select (check) this check box to take the Cleanse Match Server
offline, making it temporarily unavailable. Once offline, no
cleanse jobs are sent to that Cleanse Match Server (servlet).
• Clear (uncheck) this check box to make an offline Cleanse Match
Server available again so that Informatica MDM Hub can once
again send cleanse jobs to that Cleanse Match Server.
Note: Informatica MDM Hub looks at this field but does not set it.
Taking a Cleanse Match Server offline is an administrative action.
Thread Overrides the default thread count. The default, recommended,
Count value is 1 thread. Thread counts can be changed without needing to
restart the server. Consider the following factors:
• Number of processor cores available on your machine. You
might consider setting the number of threads to the number of
processor cores available on your machine. For example, set the
number of threads for a dual-core machine to two threads, and
set the number of threads for a single quad-core to four threads.
• Remote database connection. If you are working with a remote
database, you might consider setting the threads to a number
that is slightly higher than the number of processor cores, so
that the wait of one thread can be used by another thread.
• Process memory requirements. If you are running a memory-
intensive process, you must restrict the total memory allocated
to all threads that are run under the JVM to 1 Gigabyte. Because
Informatica MDM Hub runs in a 32-sit JVM environment, each
thread requires memory from the same JVM, and therefore the
total amount of memory is restricted.
If you set this to any illegal values (such as a negative number, 0, a
character, or a string), then it will automatically reset to the default
value (1).
Note: You must change this value after migration from an earlier
hub version or all values will default to one (1) thread.
CPU Specifies the relative CPU performance of the machines in the
Rating cleanse server pool. This value is used when deciding how to
distribute work during distributed job processing. If all of the
machines are the same, this number should remain set to the
default (1). However, if one machine’s CPU were twice as powerful
as the others, for example, consider setting this rating to 2.
- 311 -
Adding a New Cleanse Match Server
To add a new Cleanse Match Server:
1. Start the Cleanse Match Server tool. For more information, see "Starting
the Cleanse Match Server Tool" on page 310.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. In the right pane of the Cleanse Match Server tool, click the button to
add a new Cleanse Match Server.
The Cleanse Match Server tool displays the Add/Edit Match Cleanse Server
dialog
4. Set the properties for this new Cleanse Match Server. For more
information, see "Cleanse Match Server Properties" on page 310.
If proxy users have been configured for your Informatica MDM Hub
implementation, see "Cleanse Match Servers and Proxy Users" on page
309.
5. Click OK.
- 312 -
1. Start the Cleanse Match Server tool. For more information, see "Starting
the Cleanse Match Server Tool" on page 310.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Select the Cleanse Match Server that you want to configure.
5. Change the properties you want for this Cleanse Match Server. For more
information, see "Cleanse Match Server Properties" on page 310.
If proxy users have been configured for your Informatica MDM Hub
implementation, see "Cleanse Match Servers and Proxy Users" on page
309.
6. Click OK to apply your changes.
- 313 -
3. Select the Cleanse Match Server that you want to delete.
If there was a problem, Informatica MDM Hub will display a window with
information about the connection problem.
4. Click OK.
In Informatica MDM Hub, you can build and execute cleanse functions that
cleanse data. A cleanse function is a function that is applied to a data value in
- 314 -
a record to standardize or verify it. For example, if your data has a column for
salutation, you could use a cleanse function to standardize all instances of
“Doctor” to “Dr.” You can apply cleanse functions successively, or simply
assign the output value to a column in the staging table.
In Informatica MDM Hub, each cleanse function is one of the following types:
• a Informatica MDM Hub-defined function
• a function defined by your cleanse engine
• a custom cleanse function you define
Libraries
Functions are organized into libraries—Java libraries and user libraries, which
are folders used to organize the functions that you can use in the Cleanse
Functions tool in the Model workbench. For more information, see
"Configuring Cleanse Libraries" on page 317.
The functions you see in the Hub Console depend on the cleanse engine that
you are using. Informatica MDM Hub shows the cleanse functions that your
cleanse engine makes available. Regardless of which cleanse engine you use,
the overall process of data cleansing in Informatica MDM Hub is the same.
- 315 -
• In the Hub Console, expand the Model workbench and then click Cleanse
Functions.
The functions you see in the left pane depend on the cleanse engine you are
using. Your functions may differ from the ones shown in the previous figure.
Cleanse functions are grouped in the tree according to their type. Cleanse
function types are high-level categories that are used to group similar cleanse
functions for easier management and access.
If you expand the list of cleanse function types in the navigation pane, you can
select a cleanse function to display its particular properties.
- 316 -
Field Description
Read Allows a map to look up records directly from a database table.
Database Note: This function is designed to be used when there are many
references to the same limited number of data items.
Reject Allows the creator of a map to identify incorrect data and reject the
record, noting the reason.
You can add a User Library when you want to create a customized cleanse
function from existing internal or external Informatica cleanse functions.
- 317 -
1. Start the Cleanse Functions tool according to the instructions in "Starting
the Cleanse Functions Tool" on page 315.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click Refresh to refresh your cleanse library.
4. In the tree, select the Cleanse Functions node.
5. Right-click and choose Add User Library from the pop-up menu.
The Cleanse Functions tool displays the Add User Library dialog.
Field Description
Name Unique, descriptive name for this library.
Description Optional description of this library.
7. Click OK.
The Cleanse Functions tool displays the new library you added in the list
under Cleanse libraries in the navigation pane.
- 318 -
6. Specify the JAR file for this library. You can click the Browse button to
look for the JAR file.
7. Specify the following properties:
Field Description
Name Unique, descriptive name for this library.
Description Optional description of this library.
8. If applicable, click the Parameters button to specify any parameters for
this library.
The Cleanse Functions tool displays the Parameters dialog.
The name, value pairs that are imported from the file will be available
to the user-defined Java function at run time as elements of its Java
- 319 -
properties. This allows you to provide customized values in a generic
function, such as “userid” or “target URL”.
9. Click OK.
The Cleanse Functions tool displays the new library in the list under
Cleanse libraries in the navigation pane.
To learn about adding graph functions to your library, see "Configuring Graph
Functions" on page 321.
Field Description
Name Unique, descriptive name for this regular expression function.
Description Optional description of this regular expression function.
- 320 -
5. Click OK.
The Cleanse Functions tool displays the new regular expression function
under the user library in the list in the left pane, with the properties in the
right pane.
- 321 -
About Graph Functions
In Informatica MDM Hub, a graph function is a cleanse function that you can
visualize and configure graphically using the Cleanse Functions tool in the Hub
Console. You can add any pre-defined functions to a graph function.
Alternatively, to define a regular expression function, see "Configuring
Regular Expression Functions" on page 320.
For each graph function, you must configure all required inputs and outputs.
Inputs and outputs have the following properties.
Field Description
Name Unique, descriptive name for this input or output.
Description Optional description of this input or output.
Data Type Data type. Must match exactly. One of the following values:
• Boolean—accepts Boolean values only
• Date—accepts date values only
• Float—accepts float values only
• Integer—accepts integer values only
• String—accepts any data
- 322 -
Field Description
Name Unique, descriptive name for this graph function.
Description Optional description of this graph function.
5. Click OK.
The Cleanse Functions tool displays the new graph function under the
library in the list in the left pane, with the properties in the right pane.
You can add as many functions as you want to a graph function. The example
in this section shows adding only a single function.
If you already have graph functions defined, you can treat them just like any
other function in the cleanse libraries. This means that you can add a graph
function inside another graph function. This approach allows you to reuse
functions.
The area in this tab is referred to as the workspace. You might need to
resize the window to see both the input and output on the workspace.
- 323 -
By default, graph functions have one input and one output that are of type
string (gray circle). The function that you are defining might require more
inputs and/or outputs and different data types. For more information, see
"Configuring Inputs" on page 328 and "Configuring Outputs" on page 329.
4. Right-click on the workspace and choose Add Function from the pop-up
menu.
For more on the other commands on this pop-up menu, see "Workspace
Commands" on page 327. You can also add or delete these functions using
the toolbar buttons.
The Cleanse Functions tool displays the Choose Function to Add dialog.
5. Expand the folder containing the function you want to add, select the
function to add, and then click OK.
Note: The functions that are available for you to add depend on your
cleanse engine and its configuration. Therefore, the functions that you see
might differ from the cleanse functions shown in the previous figure.
The Cleanse Functions tool displays the added function in your workspace.
- 324 -
Note: Although this example shows a single graph function on the
workspace, you can add multiple functions to a cleanse function.
To move a function, click it and drag it wherever you need it on the
workspace.
- 325 -
For more information, see "Configuring Inputs" on page 328 and
"Configuring Outputs" on page 329.
7. Mouse-over the input connector, which is the little circle on the right side
of the input box. It turns red when ready for use.
8. Click the node and draw a line to one of the function input nodes.
9. Draw a line from one of the function output nodes to the output box node.
10. Click the button to save your changes. To learn about testing your new
function, see "Testing Functions" on page 330.
- 326 -
Workspace Commands
Function Modes
You can cycle through the display modes (compact, standard, and expanded)
by double-clicking on the function.
Workspace Buttons
The toolbar on the right side of the workspace provides the following buttons.
Button Description
Save changes.
- 327 -
Button Description
Edit the function outputs.
Using Constants
Constants are useful in cases where you know that you have standardized
input. For example, if you have a data set that you know consists entirely of
doctors, then you can use a constant to put Dr. in the title. When you use
constants in your graph function, they are differentiated visually from other
functions by their grey background color.
Configuring Inputs
- 328 -
Note: Once you create an input, you cannot later edit the input to change
its type. If you must change the type of an input, create a new one of the
correct type and delete the old one.
Field Description
Name Unique, descriptive name for this parameter.
Data Type Data type of this parameter.
Description Optional description of this parameter.
8. Click OK.
Add as many inputs as you need for your functions.
Configuring Outputs
- 329 -
4. Click the Details tab.
5. Right-click on the output and choose Edit outputs.
The Cleanse Functions tool displays the Outputs dialog.
Note: Once you create an output, you cannot later edit the output to
change its type. If you must change the type of an output, create a new
one of the correct type and delete the old one.
Field Description
Name Unique, descriptive name for this parameter.
Data Type Data type of this parameter.
Description Optional description of this parameter.
7. Click OK.
Add as many outputs as you need for your functions.
Testing Functions
Once you have added and configured a graph or regular expression function, it
is recommended that you test it to make sure it is behaving as expected. This
test process mimics a single record coming into the function.
- 330 -
1. Start the Cleanse Functions tool according to the instructions in "Starting
the Cleanse Functions Tool" on page 315.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Select the cleanse function that you want to test.
4. Click the Test tab.
The Cleanse Functions tool displays the test screen.
5. For each input, specify the value that you want to test by clicking the cell
in the Value column and typing a value that complies with the data type of
the input.
• For Boolean inputs, the Cleanse Functions tool displays a true/false
drop-down list.
• For Calendar inputs, the Cleanse Functions tool displays a Calendar
button that you can click to select a date from the Date dialog.
6. Click Test.
If the test completed successfully, the output is displayed in the output
section.
- 331 -
About Conditional Execution Components
Conditional execution components are useful when, for example, you have
segmented data. Suppose a table has several distinct groups of data (such as
customers and prospects). You could create a column that indicated the group
of which the record is a member. Each group is called a segment. In this
example, customers might have C in this column. while prospects would have
P. You could use a conditional execution component to cleanse the data
differently for each segment. If the conditional value does not meet any of the
conditions you specify, then the default case will be executed.
- 332 -
The Cleanse Functions tool displays the Add Value dialog.
6. Enter a value for the condition. Using the customer and prospect example,
you would enter C or P. Click OK.
The Cleanse Functions tool displays the new condition in the list of
conditions on the left, as well as in the input box.
Add as many conditions as you require. You do need to specify a default
condition—the default case is automatically created when you create a
new conditional execution component. However, you can specify the
default case with the asterisk (*). The default case will be executed for all
cases that are not covered by the cases you specify.
7. Add as many functions as you require to process all of the conditions. For
more information, see "Adding Functions to a Graph Function" on page
323.
8. For each condition—including the default condition—draw a link between
the input node to the input of the function. In addition, draw links between
the outputs of the functions and the output of your cleanse function.
Note: You can specify nested processing logic in graph functions. For
example, you can nest conditional components within other conditional
components (such as nested case statements). In fact, you can define an
entire complex process containing many conditional tests, each one of which
contains any level of complexity as well.
- 333 -
1. Start the Cleanse Functions tool according to the instructions in "Starting
the Cleanse Functions Tool" on page 315.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click Refresh to refresh your cleanse library. Used with external cleanse
engines.
Important: You must choose Refresh after acquiring a write lock and
before processing any records. Otherwise, your external cleanse engine
will throw an error.
4. Right-click your cleanse library in the list under Cleanse Functions and
choose choose Add Cleanse List.
The Cleanse Functions tool displays the Add Cleanse List dialog.
Field Description
Name Unique, descriptive name for this cleanse list.
Description Optional description of this cleanse list.
6. Click OK.
The Cleanse Functions tool displays the details pane for the new (empty)
cleanse list on the right side of the screen.
- 334 -
Input Properties
- 335 -
Property Description
matched text in the input string.
• FALSE—Replaces the match text in the input string.
Default setting if Strip is not specified.
Note: The replaceAllOccurrences parameter determines
whether replacement or removal affects all matches in
the input string or just the first match.
defaultValue Value to use for the output if none of the cleanse list
items was found in the input string. If this property is
not specified and no match was found, then the original
input string is used as the output.
Output Properties
- 336 -
4. Change the display name and description in the right pane, if you want, by
clicking the Edit button next to a value that you want to change.
5. Click the Details tab.
The Cleanse Functions tool displays the details for the cleanse list.
7. Specify a search string, an output string, a match type, and click OK.
- 337 -
The search string is the input that you want to cleanse, resulting in the
output string.
Important: Informatica MDM Hub will search through the strings in the
order in which they are entered. The order in which you specify the items
can therefore affect the results obtained. To learn more about the types of
matches available, see "Types of String Matches" on page 338.
Note: As soon as you add strings to a cleanse list, the cleanse list is
saved.
The strings that you specified are shown in the Cleanse List Details section.
8. You can add and remove strings. You can also move string forward or
backward in the cleanse list, which affects their order in run-time
execution sequence and, therefore, the results obtained.
9. You can also specify the “Default value” for every input string that does not
match any of the search strings.
If you do not specify a default value, every input string that does not
match a search string is passed to the output string with no changes.
For the output string, you can specify one of the following match types:
Match Description
Type
Exact Text string (for example, “IBM”). Note that string matches are not
Match case sensitive. For example, the string test will also match TEST or
Test.
Regular Pattern using the Java syntax for regular expressions (for
Expression example, “I.M.*” would match “IBM”, “IB Corp” and “IXM Inc.”) To
parse a name field that consists of first, middle, and last names,
you could use the following regular expression (\S+$) will give
you the last name no matter what name you give it.
The regular expression that is typed in as a parameter will be used
against the string and the matched output will be sent to the
outlet. You can also specify the group number to match an inner
group of the regular expression. Refer to the Javadoc for
java.util.regex.Pattern for the documentation on the regular
expression construction and how groups work.
SQL Match Pattern using the SQL syntax for the LIKE operator in SQL (for
example, “I_M%” would match “IBM”, “IBM Corp” and “IXM Inc.”).
- 338 -
2. Specify the connection properties for the source of the data and click
Next.
The Cleanse Functions tool displays a list of tables available for import.
- 339 -
You can import the records of the sample data either as phrases (one entry
for each record) or as words (one entry for each word in each record).
Choose whether to import the match strings as words or phrases and then
click Finish.
The Cleanse List Details box is now populated with data from the specified
source.
Note: The imported match strings are not part of the match list. To add
them to the match list, you need to move them to the Search Strings on
the right hand side.
- 340 -
• To add match strings to the match list with the match string value in
both the Search String and Output String, select the strings in the
Match Strings list, and
• To add all Match Strings to the match list, click the button.
• To clear all Match Strings from the match list, click the button.
• Repeat these steps until you have constructed a complete match list.
5. When you have finished changing the match list properties, click the
button to save your changes.
- 341 -
6. Select the columns that you want to import.
7. Click Next.
The Cleanse Functions tool displays a list of match strings available for
import.
8. Click Finish.
The Cleanse List Details box is now populated with data from the specified
source.
9. When you have finished changing the match list properties, click the
button to save your changes.
- 342 -
Chapter 13: Configuring the Load
Process
This chapter explains how to configure the load process in your Informatica
MDM Hub implementation. For an introduction, see "Load Process" on page
227.
Chapter Contents
• "Before You Begin" on page 343
• "Configuration Tasks for Loading Data" on page 343
• "Configuring Trust for Source Systems" on page 344
• "Configuring Validation Rules" on page 353
- 343 -
• "Configuring Validation Rules" on page 353
For additional configuration settings that can affect the load process, see:
• "Loading by RowID" on page 296
• "Distinct Systems" on page 445
• "Generate Match Tokens on Load" on page 92
• "Load Process" on page 227
About Trust
Several source systems may contain attributes that correspond to the same
column in a base object table. For example, several systems may store a
customer’s address. However, one system might be a more reliable source for
that data than others. If these systems disagree, then Informatica MDM Hub
must decide which value is the best one to use.
To help with comparing the relative reliability of column data from different
source systems, Informatica MDM Hub allows you to configure trust for a
column. Trust is a designation the confidence in the relative accuracy of a
particular piece of data. For each column from each source, you can define a
trust level represented by a number between 0 and 100, with zero being the
least trustworthy and 100 being the most trustworthy. By itself, this number
has no meaning. It becomes meaningful only when compared with another
trust number to determine which is higher.
Trust takes into account the age of data, how much its reliability has decayed
over time, and the validity of the data. Trust is used to determine survivorship
(when two records are consolidated), and whether updates from a source
system are sufficiently reliable to update the master record.
Trust Levels
A trust level is a number between 0 and 100. By itself, this number has no
meaning. It has meaning only when compared with another trust number.
The reliability of data from a given source system can decay (diminish) over
time. In order to reflect this fact in trust calculations, Informatica MDM Hub
allows you to configure decay characteristics for trust-enabled columns. The
- 344 -
decay period is the amount of time that it takes for the trust level to decay
from the maximum trust level (see "Maximum Trust" on page 347) to the
minimum trust level (see "Minimum Trust" on page 347). For more
information, see "Units" on page 348, "Decay" on page 348, and "Graph Type"
on page 348.
Trust Calculations
The load process calculates trust for trust-enabled columns in the base object.
For records with trust-enabled columns, the load process assigns a trust score
to cell data. This trust score is initially based on the configured trust settings
for that column. The trust score may be subsequently downgraded when the
load process applies validation rules—if configured for a trust-enabled
column—after the trust calculations. For more information, see "Run-time
Execution Flow of the Load Process" on page 231.
During the load process, if a record in the staging table will be used for a load
update operation, and if that record contains a changed cell value in a trust-
enabled column, the load process calculates trust scores for:
• the cell data in the source record in the staging table (which contains the
updated information)
• the cell data in the target record in the base object (which contains the
existing information)
If the cell data in the source record has a higher trust score than the cell data
in the target record, then Informatica MDM Hub updates the cell in the base
object record with the cell data in the staging table record.
When two records in a base object are consolidated, Informatica MDM Hub
calculates the trust score for each trusted column in the two records being
merged. Cells with the highest trust scores survive in the final consolidated
record. If the trust scores are the same, then Informatica MDM Hub compares
records according to an order of precedence, as described in "Survivorship
and Order of Precedence" on page 221.
- 345 -
For each trust-enabled column in a base object record, Informatica MDM Hub
maintains a record in a corresponding control table that contains the last
update date and an identifier of the source system. Based on these settings,
Informatica MDM Hub can always calculate the current trust for the column
value.
If history is enabled for a base object, Informatica MDM Hub also maintains a
separate history table for the control table, in addition to history tables for the
base object and its cross-reference table.
The cross-reference table for a base object contains the most recent value
from each source system. By default (without trust settings), the base object
contains the most recent value no matter which source system it comes from.
For trust-enabled columns, the cell value in a base object record might not
have the same value as its corresponding record in the cross-reference table.
Validation rules, which are run during the load process after trust calculations,
can downgrade trust for a cell so that a source that had previously provided
the cell value might not update the cell. For more information about validation
rules, see "Configuring Validation Rules" on page 353.
- 346 -
Overriding Trust Scores
Data stewards can manually override a calculated trust setting if they have
direct knowledge that a particular value is correct. Data stewards can also
enter a value directly into a record in a base object. For more information, see
the Informatica MDM Hub Data Steward Guide.
For state-enabled base objects, trust is calculated for records with a PENDING
or ACTIVE state, but records with a DELETE state are ignored. For more
information, see "State Management" on page 159.
Synchronize batch jobs can fail for base objects with a large number of trust-
enabled columns. Similarly, Automerge jobs can fail if there is a large number
of trust-enabled or validation-enabled columns. The exact number of columns
that cause the job to fail is variable and is based on the length of the column
names and the number of trust-enabled columns (or, for Automerge jobs,
validation-enabled columns as well). Long column names are at—or close to—
the maximum allowable length of 26 characters. To avoid this problem, keep
the number of trust-enabled columns below 60 and/or the length of the column
names short. A work around is to enable all trust/validation columns before
saving the base object to avoid running the synchronization job.
Trust Properties
This section describes the trust properties that you can configure for trust-
enabled columns. Trust properties are configured separately for each source
system that could provide records for trust-enabled columns in a base object.
Maximum Trust
The maximum trust (starting trust) is the trust level that a data value will
have if it has just been changed. For example, if source system X changes a
phone number field from 555-1234 to 555-4321, the new value will be given
system X’s maximum trust level for the phone number field. By setting the
maximum trust level relatively high, you can ensure that changes in the
source systems will usually be applied to the base object.
Minimum Trust
The minimum trust is the trust level that a data value will have when it is old
(after the decay period has elapsed). This value must be less than or equal to
- 347 -
the maximum trust.
Note: If the maximum and minimum trust are equal, then the decay curve is
a flat line and the decay period and decay type have no effect.
Units
Specifies the units used in calculating the decay period—day, week, month,
quarter, or year.
Decay
Specifies the number (of days, weeks, months, quarters, or years) used in
calculating the decay period.
Note: For the best graph view, limit the decay period you specify to between
1 and 100.
Graph Type
Decay follows a pattern in which the trust level decreases during the decay
period. The graph types show these decay patterns have any of the following
settings.
Icon Graph Description
Type
Linear Simplest decay. Decay follows a straight line from the
maximum trust to the minimum trust.
By default, the start date for trust decay shown in the Trust Decay Graph is the
current system date. To see the impact of trust decay based on a different
- 348 -
start date for a given source system, specify a different test offset date
according to the instructions in "Changing the Offset Date for a Trust-Enabled
Column" on page 352.
Trust is disabled by default. When trust is disabled, Informatica MDM Hub uses
the value from the most recently-executed load process regardless of which
source system it comes from. If column data for a base object comes from
only one system, then trust should remain disabled for that column.
Trust should be enabled, however, for columns in which data can come from
multiple source systems. If you enable trust for a column, you also assign
trust levels to specify the relative reliability of any source systems that could
provide records that update the column.
- 349 -
Before You Configure Trust for Trust-Enabled Columns
Before you configure trust for trust-enabled columns, you must have:
• enabled trust for base object columns according to the instructions in
"Enabling Trust for a Column" on page 349
• configured staging tables in the Schema Manager, including associated
source systems and staging table columns that correspond to base object
columns, according to the instructions in "Configuring Staging Tables" on
page 275
- 350 -
The Systems and Trust tool displays a read-only view of the trust-enabled
columns in the selected base object, indicating with a check mark whether
a given source system supplies data for that column.
Note: The association between trust-enabled columns and source systems
is specified in the staging tables for this base object. For more
information, see "Configuring Staging Tables" on page 275.
5. Expand a base object to see its trust-enabled columns.
The X-axis is the trust score and the Y-axis is the time.
- 351 -
Changing the Offset Date for a Trust-Enabled Column
By default, the Trust Decay Graph shows the trust decay across all source
systems from the current system date. You can specify a different date (such
as a future date) to test your current trust settings and see how trust would
decay from that date. Note that offset dates are not saved.
2. Click the Calendar button next to the source system for which you want
to specify a different offset date.
The Systems and Trust tool prompts you to specify a date.
After records have been loaded into a base object, if you enable trust for any
column, or if you change trust settings for any trust-enabled column(s) in that
- 352 -
base object, then you must run the Synchronize batch job (see "Synchronize
Jobs" on page 557) before running the consolidation process. If this batch job
is not run, then errors will occur during the consolidation process.
consists of:
• Condition: Length < 3
• Action: Downgrade trust on First_Name by 50%
If the Reserve Minimum Trust flag is set for the column, then the trust cannot
be downgraded below the column’s minimum trust. You use the Schema
Manager to configure validation rules for a base object.
Validation rules are executed during the load process, after trust has been
calculated for trust-enabled columns in the base object. If validation rules
have been defined, then the load process applies them to determine the final
trust scores, and then uses the final trust values to determine whether to
update records in the base object with cell data from the updated records. For
more information, see "Run-time Execution Flow of the Load Process" on page
231.
Validation Checks
- 353 -
For example, supposed you used an address verification flag in which the flag
is OK if the address is complete and BAD if the address is not complete. You
could configure a validation rule that downgrades the trust on all address
fields if the verification flag is not OK. Note that, in this case, the verification
flag should also be downgraded.
Required Columns
Validation rules are applied regardless of the source of the incoming data.
However, validation rules are applied only if the staging table or if the input—
a Services Integration Framework (SIF) request—contains all of the required
columns. If any required columns are missing, validation rules are not
applied.
If a base object contains existing data and you change validation rules, you
must run the Revalidate job to recalculate trust scores for new and existing
data, as described in "Revalidate Jobs" on page 556.
For state-enabled base objects, validation rules are applied to records with a
PENDING or ACTIVE state, but records with a DELETE state are ignored. For
more information, see "State Management" on page 159
- 354 -
Validation rules are disabled by default. Validation rules should be enabled,
however, for any trust-enabled columns that will use validation rules for trust
downgrades.
Note: The execution sequence for validation rules differs between the load
process described in this chapter and PUT requests invoked by external
applications using the Services Integration Framework (SIF). For PUT
requests, validation rules are executed in order of decreasing downgrade
percentage. For more information, see the Informatica MDM Hub Services
Integration Framework Guide and the Informatica MDM Hub Javadoc.
- 355 -
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Expand the tree for the base object that you want to configure, and then
click its Validation Rules Setup node.
The Schema Manager displays the Validation Rules editor.
Pane Description
Number Number of configured validation rules for the selected base
of Rules object.
Validation List of configured validation rules for the selected base object.
Rules
Properties Properties for the selected validation rule. For more
Pane information, see "Validation Rule Properties" on page 356.
Rule Name
Rule Type
- 356 -
Rule Columns
For each column, you specify the downgrade percentage and whether to
reserve minimum trust.
Downgrade Percentage
Percentage by which the trust level of the specified column will be decreased
if this validation rule condition is met. The larger the percentage, the greater
the downgrade. For example, 0% has no effect on the trust, while 100%
downgrades the trust completely (unless the reserve minimum trust is
specified, in which case 100% downgrades the trust so that it equals minimum
trust).
If trust is downgraded by 100% and you have not enabled minimum reserve
trust for the column, then the value of that column will not be populated into
the base object.
Specifies what will happen if the downgrade causes the trust level to fall below
the column’s minimum trust level. You can retain the minimum trust (so that
the trust level will be reduced to the minimum trust but no lower). If this box
is cleared (unchecked), then the trust level will be reduced by the specified
percentage even if this means going below the minimum trust.
Rule SQL
Specifies the SQL WHERE clause representing the condition for this validation
rule. During the load process, the validation rule is executed. If data meets
the criteria specified in the Rule SQL field, then the trust value is downgraded
by the downgrade percentage configured for this validation rule.
The Validation Rules editor prompts you to configure the SQL WHERE clause
based on the selected Rule Type for this validation rule.
During the load process, this query is used to check the validity of the data in
the staging table.
The following table provides examples of SQL WHERE clauses based on the
selected rule type.
- 357 -
Examples of WHERE Clause for Each Rule Type
Rule Type WHERE clause Examples Result
Existence WHERE S.ColumnName WHERE S.MIDDLE_ Affected columns will be
Check IS NULL NAME IS NULL downgraded for records
with middle names that
are null. The records that
do not meet the condition
will not be affected.
Domain WHERE S.ColumnName WHERE S.Gender NOT Affected columns will be
Check IN ('?', '?', '?') IN ('M', 'F', 'U') downgraded if the Gender
is any value other than M,
F, or U.
Referential WHERE NOT EXISTS WHERE NOT EXISTS Affected columns will be
Integrity (SELECT <blank>’a’ (SELECT DISTINCT downgraded for records
FROM ? WHERE ?.? = 'a' FROM ACCOUNT_ with Account Type values
S.<Column_Name> TYPE WHERE
that are not on the
Account Type table.
WHERE NOT EXISTS ACCOUNT_
(SELECT <blank> TYPE.Account_Type
'a' FROM <Ref_ = S.Account_Type
Table> WHERE <Ref_
Table>.<Ref_
Column> =
S.<Column_Name>
Pattern WHERE S.ColumnName WHERE S.eMail_ Downgrade will be applied
Validation LIKE 'Pattern' Address NOT LIKE if the e-mail address does
'%@%' not contain an @
character.
Custom WHERE WHERE Downgrade will be applied
LENGTH(S.ZIP_CODE) if the length of the zip
> 4 code column is less than
4.
You can use the wildcard character (*) to reference tables via an alias.
• s.* aliases the staging table
• I.* aliases a temporary table and provides ROWID_OBJECT, PKEY_SRC_
OBJECT, and ROWID_SYSTEM information for the records being updated.
For Custom rule types, write SQL statements that are well formed and well
tuned. If you need more information about SQL WHERE clause syntax and wild
card patterns, refer to the product documentation for the database platform
used in your Informatica MDM Hub implementation.
- 358 -
WHERE conditionA OR conditionB or conditionC
These two statements will yield very different results when evaluating
records.
3. Specify the properties for this validation rule. For more information, see
"Validation Rule Properties" on page 356.
- 359 -
4. If you want, select the rule column(s) for this validation rule by clicking
the button.
The Validation Rules editor displays the Select Rule Columns dialog.
The available columns are those that have the Validate flag enabled (see
"Column Properties" on page 103. For more information, see "Configuring
Columns in Tables" on page 102.
Select the column(s) for which the trust level will be downgraded if the
condition specified in the WHERE clause for this validation rule is met, and
then click OK.
5. Click OK.
The Schema Manager adds the new rule to the list of validation rules.
Note: If a base object contains existing data and you change validation
rules, you must run the Revalidate job to recalculate trust scores for new
and existing data, as described in "Revalidate Jobs" on page 556.
- 360 -
3. Specify the editable properties for this validation rule. You cannot change
the rule type. For more information, see "Validation Rule Properties" on
page 356.
4. If you want, select the rule column(s) for this validation rule by clicking
the button.
The Validation Rules editor displays the Select Rule Columns dialog.
The available columns are those that have the Validate flag enabled (see
"Column Properties" on page 103. For more information, see "Configuring
Columns in Tables" on page 102.
Select the column(s) for which the trust level will be downgraded if the
condition specified in the WHERE clause for this validation rule is met, and
then click OK.
- 361 -
Changing the Sequence of Validation Rules
The execution order for validation rules is extremely important. For more
information, see "Execution Sequence of Validation Rules" on page 355.
Use the following buttons to change the sequence of validation rules in the list.
Click To....
Move the selected validation rule higher in the sequence.
- 362 -
Chapter 14: Configuring the Match
Process
This chapter describes how to configure your Hub Store to identify and handle
potential duplicate records. For an introduction to the match process, see
"Match Process" on page 245.
Chapter Contents
• "Configuration Tasks for the Match Process" on page 363
• "Navigating to the Match/Merge Setup Details Dialog" on page 365
• "Configuring Match Properties for a Base Object" on page 366
• "Configuring Match Paths for Related Records" on page 373
• "Configuring Match Columns" on page 387
• "Configuring Match Rule Sets" on page 399
• "Configuring Match Column Rules for Match Rule Sets" on page 407
• "Configuring Primary Key Match Rules" on page 434
• "Investigating the Distribution of Match Keys" on page 438
• "Excluding Records from the Match Process" on page 441
- 363 -
Understanding Your Data
Before you define match rules, you must be very familiar with your data and
understand:
• the distribution of the values in the columns you intend to use to determine
duplicate records, and
• the general proportion of the total number of records that are duplicates.
- 364 -
8. Tune your rules. This is an iterative process by which you apply your
match rules to a representative data set, analyze the results, and adjust
your settings to optimize the match performance.
If you want to change settings, you need to Acquire a write lock according
to the instructions in "Acquiring a Write Lock" on page 36.
- 365 -
Tab Name Description
"Configuring Match Properties for a Base Object" on page 366.
Paths Allows you to configure the match path for parent/child
relationships for records in different base objects or in the same
base object. For more information, see "Configuring Match Paths
for Related Records" on page 373.
Match Allows you to configure match columns for match column rules.
Columns For more information, see "Configuring Match Columns" on page
387 and "Configuring Match Column Rules for Match Rule Sets" on
page 407.
Match Rule Allows you to define a search strategy and rules using match rule
Sets sets. For more information, see "Configuring Match Rule Sets" on
page 399.
Primary Allows you to define primary key match rules. For more
Key Match information, see "Configuring Primary Key Match Rules" on page
Rules 434.
Match Key Shows the distribution of match keys. For more information, see
Distribution "Investigating the Distribution of Match Keys" on page 438.
Merge Allows you to merge and link settings. For more information, see
Settings "Configuring the Consolidate Process" on page 443
- 366 -
3. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
Match Properties
This section describes the configuration settings on the Match Properties tab.
- 367 -
Calculated, Read-Only Fields
This setting helps prevent data stewards from being overwhelmed with
thousands of matches for manual consolidation. This sets the limit on the list
of possible matches that must be decided upon by a data steward (default is
1000). Once this limit is reached, Informatica MDM Hub stops the match
process until the number of records for manual consolidation has been
reduced.
This setting specifies an upper limit on the number of records that Informatica
MDM Hub will process for matching during match process execution (Match or
Auto Match and Merge jobs). When the match process starts executing, it
begins by flagging records to be included in the match job batch. From the
pool of new/unconsolidated records that are ready for match
(CONSOLIDATION_IND=4, as described in "Consolidation Indicator" on page
219), the match process changes CONSOLIDATION_IND to 3. The number of
records flagged is determined by the Number of Rows per Match Job Batch
Cycle. The match process then matches those records in the match job batch
against all of the records in the base object.
The number of records in the match job batch affects how long the match
process takes to execute. The value to specify depends on the size of your
data set, the complexity of your match rules, and the length of the time
window you have available to run the match process. The default match batch
size is low (10). You increase this based on the number of records in the base
- 368 -
object, as well as the number of matches generated for those records based
on its match rules.
• The lower your match batch size, the more times you will need to run the
match and consolidation processes.
• The higher your match batch size, the more work each match and
consolidation process does.
For each base object, there is a medium ground where you reach the optimal
match batch size. You need to identify this optimal batch size as part of
performance tuning in your environment. Start with a match batch size of 10%
of the volume of records to be matched and merged, run the match job only,
see how many matches are generated by your match rules, and then adjust
upwards or downwards accordingly.
Enable (set to Yes) this feature to have Informatica MDM Hub mark as unique
(CONSOLIDATION_IND=1) any records that have been through the match
process, but for which no matches were identified. If enabled, for such
records, Informatica MDM Hub automatically changes their state to
consolidated (changes the consolidation indicator from 2 to 1). Consolidated
records are removed from the data steward’s queue via the Automerge batch
job.
- 369 -
Match/Search Strategy
Select the match/search strategy to specify the reliability of the match versus
the performance you require. Select one of the following options.
Strategy Description
Option
Fuzzy Probabilistic match that takes into account spelling variations,
possible misspellings, and other differences that can make matching
records non-identical. This is the primary means of matching data in
a base object. Referred to in this document as fuzzy-match base
objects.
Note: If you specify a Fuzzy match/search strategy, you must
specify a fuzzy match key.
Exact Matches only records with identical values in the match column(s). If
you specify an exact match, you can define only exact-match
columns for this base object (exact-match base objects cannot have
fuzzy-match columns). Referred to in this document as exact-match
base objects.
An exact strategy is faster, but an exact match will miss some matches if the
data is imperfect. The best option to choose depends on the characteristics of
the data, your knowledge of the data, and your particular match and
consolidation requirements.
Certain configuration settings the Match / Merge Setup tab apply to only one
type of base object. In this document, such features are indicated with a
graphic that shows whether it applies to fuzzy-match base objects only (as in
the following example), or exact-match base objects only. No graphic means
that the feature applies to both.
Note: The match / search strategy is configured at the base object level. For
more information about the match / search strategy configured at the match
rule level, see "Match / Search Strategy" on page 409.
Fuzzy Population
- 370 -
populations per country. If you require another population, contact
Informatica support. If you chose an exact match/search strategy, then this
value is ignored.
If this setting is enabled (checked), then Informatica MDM Hub matches the
current records against records with lower ROWID_OBJECT values. For
example, if the current record has a ROWID_OBJECT value of 100, then the
record will be matched only against other records in the base object with a
ROWID_OBJECT value that is less than 100 (ignoring all records with a
ROWID_OBJECT value that is higher than 100).
Using this feature can reduce the number of matches required and speed
performance. However, if PUTs are executed, or if records are inserted out of
rowid order, then records might not be fully matched. You must assess the
trade-off between performance and match quantity based on the
characteristics of your data and your particular match requirements. By
default, this option is disabled (unchecked).
Available only for fuzzy key matching and only if “"Match Only Previous Rowid
Objects" on page 371” is checked (selected). If Match Only Once is enabled
(checked), then once a record has found a match, Informatica MDM Hub will
not match it any further within this search range (the set of similar match key
values). Using this feature can reduce duplicates and increase performance.
- 371 -
Instead of finding every match for a record in a search range, Informatica
MDM Hub can find a single match for each. In subsequent match cycles, the
merge process will put these into large groups of XREF records associated
with the base object.
During the match process, dynamic match analysis determines whether the
match process will take an unacceptably long period of time. This threshold
value specifies the maximum acceptable number of comparisons.
To enable the dynamic match threshold, specify a non-zero value. Enable this
feature if you have data that is very similar (with high concentrations of
matches) to reduce the amount of work expended for a hot spot in your data.
A hotspot is a group of records representing overmatched data—a large
intersection of matches. If Dynamic Match Analysis Threshold is enabled, then
records that produce more than the specified number of potential match
candidates will be skipped during the match process. By default, this option is
zero (disabled).
By default, the match process includes only ACTIVE records and ignores
PENDING records. For state management-enabled objects, select this check
box to include PENDING records in the match process. Note that, regardless of
this setting, DELETED records are ignored by the match process. For more
information, see "Enabling Match on Pending Records" on page 163.
For link-style base objects only, you can unlink consolidated records and
requeue them for match. This can be configured to occur automatically on load
- 372 -
update, or manually by via the Reset Links batch job. For more information,
see "Reset Links Jobs" on page 555.
For link-style base objects only, the Schema Manager displays the following
properties.
Property Description
Allow prompt for reset of Specifies whether to prompt for a reset of match
match links when match links when configuration settings for match rules
rules / columns are or match columns are changed.
changed
Allow reset of match links Specifies whether the reset links prompt applies
for updated data to updated data (load updates). This prompt is
triggered automatically upon load update.
Allow reset of links to Specifies whether the reset links process applies
include consolidated to consolidated records.
records Note: The reset links process always applies to
unconsolidated records.
Allow reset of links to Specifies whether manually-linked records are
include manually linked included by the reset links process. Autolinked
records records are always included.
Note: This setting affects the scope of all other
reset links settings.
Match Paths
- 373 -
single base object (intra-table paths). Match paths are used for configuring
match column rules involving related records in either separate tables or in
the same table.
Configuring match paths that point to other records involves two main
components:
Component Description
foreign key Used to traverse the relationships to other records. Allows you
relationships to specify parent-to-child and child-to-parent relationships.
filters Allow you to selectively include or exclude records based on
(optional) values in a given column, such as ADDRESS_TYPE or PARTY_
TYPE. For more information, see "Configuring Filters for Match
Paths" on page 384.
Inter-Table Paths
- 374 -
In some cases, however, the relationship between records can be more
complex, requiring an intermediary base object that defines the relationship
between records in the two tables.
In order to configure match rules for this kind of relationship between records
in different base objects, you would create a separate base object (such as
PersAddrRel) that describes to Informatica MDM Hub the relationships
between records in the two base objects.
- 375 -
Column Type Description
1
ADDRESS_ VARCHAR(50) Second address line.
2
CITY VARCHAR(50) City
STATE_ VARCHAR(50) State or province
PROV
POSTAL_ VARCHAR(50) Postal code
CODE
... ... ...
To define the relationship between records in the two base objects, the
PersonAddrRel base object could have the following columns:
Column Type Description
ROWID_ CHAR(14) Primary key. Uniquely identifies this person in the
OBJECT base object.
PERS_FK CHAR(14) Foreign key to the ROWID_OBJECT column in the
Person base object.
ADDR_FK CHAR(14) Foreign key to the ROWID_OBJECT column in the
Address base object.
After you have configured the relationship base object (PersonAddrRel), you
would complete the following tasks:
1. Configure foreign keys from this base object to the ROWID_OBJECT of the
Person and Address base objects. For more information, see "Configuring
Foreign-Key Relationships Between Base Objects" on page 113.
2. Load the PersAddrRel base object with data that describes the
relationships between records.
- 376 -
ROWID_OBJECT PERS_FKEY ADDR_FKEY
2 480 920
3 786 432
4 786 980
5 12 1028
6 922 1028
7 1302 110
... ... ...
In this example, note that Person #786 has two addresses, and that
Address #1028 has two persons.
3. Use the PersonAddrRel base object when configuring match column rules
for the related records. For more information, see "Configuring Match
Column Rules for Match Rule Sets" on page 407.
Intra-Table Paths
- 377 -
The relationships among employees is hierarchical. The CEO is at the top of
the hierarchy, representing what is called the global ultimate parent record.
In order to configure match rules for this kind of object, you would create a
separate base object to describe to Informatica MDM Hub the relationships
between records. For example, you could create and configure a EmplRepRel
base object with the following columns:
Column Type Description
ROWID_ CHAR(14) Primary key. Uniquely identifies this relationship
OBJECT record.
EMPLOYEE_FK CHAR(14) Foreign key to the ROWID_OBJECT of the employee
record.
REPORTS_TO_ CHAR(14) Foreign key to the ROWID_OBJECT of a manager
FK record.
After you have configured this base object, you must complete the following
tasks:
1. Configure foreign keys from this base object to the ROWID_OBJECT of the
Employee base object. For more information, see "Configuring Foreign-
Key Relationships Between Base Objects" on page 113.
- 378 -
2. Load this base object with data that describes the relationships between
records.
Note: This example used a REPORTS_TO field to define the relationship, but
you could use piece of information to associate the records—even something
more generic and flexible like RELATIONSHIP_TYPE.
- 379 -
Navigating to the Paths Tab
To navigate to the Paths tab for a base object:
1. In the Schema Manager, navigate to the Match/Merge Setup Details dialog
for the base object that you want to configure. For more information, see
"Navigating to the Match/Merge Setup Details Dialog" on page 365.
2. Click the Paths tab.
The Schema Manager displays the Paths tab.
- 380 -
Configuring Path Components
This section describes how to configure path components in the Schema
Manager. Path components provide a way to define the connection between
parent and child tables using foreign keys for the purpose of using columns
from that table in a match column.
Display Name
The name of this path component as it will be displayed in the Hub Console.
Physical Name
Actual name of the path component in the database. Informatica MDM Hub will
suggest a physical name for the path component based on the display name
that you enter.
The Check for Missing Children check box instructs Informatica MDM Hub to
either allow for missing child records (enabled, the default) or to require all
parent records to have child records.
Setting Description
Enabled If you might have some missing child records and you have rules
(Checked) that do not include columns in the tables that might be missing
records.
Disabled If all of your rules use the child columns and do not have null
(Unchecked) match enabled. In this case, checking for missing children does
not add any value, and it can have an negative impact on
performance.
If you are certain that your data is complete (parent records have child
records), and you include the parent in the child match rule, then inter-table
matching works as expected. However, if your data tends to contain parent
records that are missing child records, or if you do not include the parent
column in the child match rule, you must check (select) the Check for Missing
Children check box in the path component associated with this match column
rule to ensure that an outer join occurs when Informatica MDM Hub checks for
records to match.
- 381 -
Note: If the Check for Missing Children option is enabled, Informatica MDM
Hub performs an outer join between the parent and child tables, which can
have a performance impact. Therefore, when not needed, it is more efficient
to disable this option.
Constraints
Property Description
Table List of tables in the schema.
Direction Direction of the foreign key:
• Parent-to-Child
• Child-to-Parent
• N/A
Foreign Column to which the foreign key points. This column can be either in
Key On a different base object or the same base object.
- 382 -
4. Specify the properties for this path component. For more information, see
"Properties of Path Components" on page 381.
5. Click OK.
5. Specify the properties for this path component. You can change the
following values:
• Display Name (see "Display Name" on page 381)
• Check for Missing Children (see "Check For Missing Children" on page
381)
6. Click OK.
You can delete path components but not the root base object. To delete a path
component:
1. In the Schema Manager, navigate to the Paths tab according to the
instructions in "Navigating to the Paths Tab" on page 380.
- 383 -
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. In the Path Components tree, select the path component that you want to
delete.
About Filters
Filter Properties
Example Filter
- 384 -
Setting Example Value
Operator IN
Values MAILING
Adding Filters
If you add multiple filters, Informatica MDM Hub evaluates the entire
expression using the logical AND operator. For example,
xExpr AND yExpr AND zExpr
To add a filter:
1. In the Schema Manager, navigate to the Paths tab according to the
instructions in "Navigating to the Paths Tab" on page 380.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
4. Specify the properties for this path component. For more information, see
"Properties of Path Components" on page 381.
5. Specify the value(s) for this filter according to the instructions in "Editing
Values for a Filter" on page 385.
- 385 -
• Edit filter properties. For more information, see "Editing Filter
Properties" on page 386.
2. In either the Add Filter or Edit Filter dialog, click the button next to the
Values field.
The Schema Manager displays the Edit Values dialog.
3. Configure the values for this filter.
4. Specify the properties for this path component. For more information, see
"Properties of Path Components" on page 381.
5. Specify the value(s) for this filter according to the instructions in "Editing
Values for a Filter" on page 385.
- 386 -
6. Click the button to save your changes.
Deleting Filters
To delete a filter:
1. In the Schema Manager, navigate to the Paths tab according to the
instructions in "Navigating to the Paths Tab" on page 380.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. In the Filters section, select the filter that you want to delete, and then
click the button.
The Schema Manager prompts you to confirm deletion.
4. Click Yes.
- 387 -
Match Columns Depend on the Search Strategy
The types of match columns that you can configure depend on the type of the
base object that you are configuring (see "Exact-match and Fuzzy-match Base
Objects" on page 247). The type of base object is defined by the selected
match / search strategy (see "Match/Search Strategy" on page 370).
Match Description
Strategy
Fuzzy- Allows you to configure fuzzy-match columns as well as exact-
match match columns. For more information, see "Configuring Match
base Columns for Fuzzy-match Base Objects" on page 390.
objects
Exact- Allows you to configure exact-match columns but not fuzzy-match
match columns. For more information, see "Configuring Match Columns
base for Exact-match Base Objects" on page 396.
objects
Path Component
The path component is either the source table to use for a match column
definition, or the match path used to navigate a hierarchy of records. Match
paths are used for configuring match column rules involving related records in
either separate tables or in the same table. Before you can specify a path
component, the match path must be configured. For more information, see
"Configuring Match Paths for Related Records" on page 373.
Field Types
For fuzzy-match columns, the field name drop-down list displays the following
field types. For more information, see "Adding Exact-match Columns for
Fuzzy-match Base Objects" on page 395.
- 388 -
Field Types
Field Name Description
Address_ Includes the part of address up to, but not including, the locality
Part1 last line. The position of the address components should be the
normal word order used in your data population. Pass this data
in one field. Depending on your base object, you may
concatenate these attributes into one field before matching. For
example, in the US, an Address_Part1 string includes the
following fields: Care-of + Building Name + Street Number +
Street Name + Street Type + Apartment Details. Address_
Part1 uses methods and options designed specifically for
addresses.
Address_ Locality line in an address. For example, in the US, a typical
Part2 Address_Part2 includes: City + State + Zip (+ Country).
Matching on Address_Part2 uses methods and options designed
specifically for addresses.
Attribute1, Two general purpose fields. These fields are matched using a
Attribute2 general purpose, string matching algorithm that compensates
for transpositions and missing characters or digits.
Date Matches any type of date, such as date of birth, expiry date,
date of contract, date of change, creation date, and so on. It
expects the date to be passed in Day+Month+Year format. It
supports the use or absence of delimiters between the date
components. Matching on dates uses methods and options
designed specifically for dates. It overcomes the typical error
and variation found in this data type.
ID Matches any type of ID number, such as: Account number,
Customer number, Credit Card number, Drivers License
number, Passport, Policy number, SSN or other identity code,
VIN, and so on. It uses a string matching algorithm that
compensates for transpositions and missing characters or
digits.
Organization_ Matches the names of organizations, such as company names,
Name business names, institution names, department names, agency
names, trading names, and so on. This field supports matching
on a single name or on a compound name (such as a legal
name and its trading style). You may also use multiple names
(for example, a legal name and a trading style) in a single
Organization_Name column for the match.
Person_Name Matches the names of people. Use the full person name. The
position of the first name, middle names, and family names,
should be the normal word order used in your population. For
example, in English-speaking countries, the normal order is:
First Name + Middle Name(s) + Family Name(s). Depending on
your base object design, you can concatenate these fields into
one field before matching. This field supports matching on a
single name, or an account name (such as JOHN & MARY
SMITH). You may also use multiple names, such as a married
name and a former name.
Postal_Area Can be used to place more emphasis on the postal code than if
it were included in the Address_Part2 field. It is for all types of
postal codes, including Zip codes. It uses a string matching
algorithm that compensates for transpositions and missing
characters or digits.
Telephone_ Used to match telephone numbers. It uses a string matching
Number algorithm that compensates for transpositions and missing
digits or area codes.
- 389 -
Selecting Multiple Columns for Matching
For example:
Anna Maria Gonzales MD
Fuzzy-match base objects can have both fuzzy and exact-match columns. For
exact-match base objects instead, see "Configuring Match Columns for Exact-
match Base Objects" on page 396.
- 390 -
The Match Columns tab for a fuzzy-match base object has the following
sections.
Property Description
Fuzzy Properties for the fuzzy match key. For more information, see
Match Key "Configuring Fuzzy Match Key Properties" on page 391.
Match Match columns and their properties:
Columns • Field Name (see "Field Types" on page 388)
• Column Type (see "Match Column Types" on page 387)
• Path Component (see "Path Component" on page 388)
• Source Table—table referenced in the path component, or
the base object (if the path component is root)
Match List of available columns in the base object, as well as
Column columns that have been selected for match.
Contents
- 391 -
This section describes how to configure the match column properties for
fuzzy-match base objects (see "Match/Search Strategy" on page 370). The
Fuzzy Match Key is a special column in the base object that the Schema
Manager adds if a match column uses the fuzzy match / search strategy. This
column is the primary field used during searching and matching to generate
match candidates for this base object. All fuzzy base objects have one and
only one Fuzzy Match Key.
Key Types
For a fuzzy-match base object, you can select one of the following key types:
Key Type Description
Person_Name Used if your fuzzy match key contains data for individuals only.
Organization_ Used if your fuzzy match key contains data for organizations
Name only, or if it contains data for both organizations and
individuals.
Address_ Used if your fuzzy match key contains address data to be
Part1 consolidated.
Note: Key types are based on the population you select. The above list of key
types applies to the default population (US). Other populations might have
different key types. If you require another population, contact Informatica
support.
Key Widths
The match key width determines the thoroughness of the analysis of the fuzzy
match key, the number of possible match candidates returned, and how much
disk space the keys consume. Key widths apply to fuzzy-match objects only.
- 392 -
Key Description
Width
Standard Appropriate for most fuzzy match keys, balancing reliability and
space usage.
Extended Might result in more match candidates, but at the cost of longer
processing time to generate keys. This option provides some
additional matching capability due to the concatenation of columns.
This key width works best when:
• your data set is not extremely large
• your data set is not complete
• you have sufficient resources to handle the processing time and
disk space requirements
Limited Trades some match reliability for disk space savings. This option
might result in fewer match candidates, but searches can be faster.
This option works well if you are willing to undermatch for faster
searches that use less disk space for the keys. Limited keys match
fewer records with word-order variations than standard keys. This
choice provides a subset of the Standard key set, but might be the
best option if disk space is restricted or the data volume is
extremely large.
Preferred Generates a single key per base object record. This option trades
some match reliability for performance (reduces the number of
matches that need to be performed) and disk space savings
(reduces the size of the match key table). Depending on
characteristics of the data, a preferred key width might result in
fewer match candidates.
Property Description
Key Type Type of field primarily used in the match. This is the main
criterion for the search that builds the initial list of potential
match candidates. This key type should be based on the main
type of data stored in the base object. For more information,
see "Key Types" on page 392.
Key Width Size of the search range for which keys are generated. For
more information, see "Key Widths" on page 392.
Path Path component for this fuzzy match key. This is a table
Component containing the column(s) to designate as the key type: Base
Object, Child Base Object table, or Cross-reference table. For
more information, see "Path Component" on page 388.
- 393 -
Adding a Fuzzy-match Column for Fuzzy-match Base Objects
Property Description
Match Path Match path component for this fuzzy-match column. For a
Component fuzzy-match column, the source table can be the parent table, a
parent cross-reference table, or any child base object table. For
more information, see "Path Component" on page 388.
Field Name of this field as it will be displayed in the Hub Console. For
Name fuzzy match columns, this is a drop-down list where you can
select the type of data in the match column being defined, as
described in "Field Types" on page 388.
5. Specify the base object column(s) for the fuzzy match.
To add a column to the Selected Columns list, select a column name and
then click the right arrow button.
Note: If you add multiple columns, the values are concatenated, with a
separator space between values. For more information, see "Selecting
Multiple Columns for Matching" on page 390.
6. Click OK.
The Schema Manager adds the match column to the Match Columns list.
- 394 -
7. Click the Save button to save your changes.
Property Description
Match Path Match path component for this exact-match column. For an
Component exact-match column, the source table can be the parent table
and / or child physical columns. For more information, see
"Path Component" on page 388.
Field Name of this field as it will be displayed in the Hub Console.
Name
5. Specify the base object column(s) for the exact match.
To add a column to the Selected Columns list, select a column name and
then click the right arrow.
Note: If you add multiple columns, the values are concatenated, with a
separator space between values. For more information, see "Selecting
Multiple Columns for Matching" on page 390.
Note: Concatenating columns is not recommended for exact match
columns.
6. Click OK.
- 395 -
The Schema Manager adds the match column to the Match Columns list.
Before you define match column rules, you must define the match columns on
which they will be based. Exact-match base objects can have only exact-
match columns. For more information about configuring match columns for
fuzzy-match base objects instead, see "Configuring Match Columns for Fuzzy-
match Base Objects" on page 390.
- 396 -
Navigating to the Match Columns Tab for an Exact-match Base
Object
The Match Columns tab for an exact-match base object has the following
sections.
Property Description
Match Match columns and their properties:
Columns • Field Name
• Column Type (see "Match Column Types" on page 387)
• Path Component (see "Path Component" on page 388)
• Source Table—table referenced in the path component, or
the base object (if the path component is root)
Match List of available columns and columns selected for matching.
Column
Contents
You can add only exact-match columns for exact-match base objects. Fuzzy-
match columns are not allowed.
- 397 -
To add an exact-match column for an exact-match base object:
1. In the Schema Manager, navigate to the Match Columns tab. For more
information, see "Navigating to the Match Columns Tab for an Exact-match
Base Object" on page 397.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
Property Description
Match Path Match path component for this exact-match column. For an exact-
Component match column, the source table can be the parent table and / or
child physical columns. For more information, see "Path
Component" on page 388.
Field Name of this field as it will be displayed in the Hub Console.
Name
5. Specify the base object column(s) for the exact match.
To add a column to the Selected Columns list, select a column name and
then click the right arrow.
Note: If you add multiple columns, the values are concatenated, with a
separator space between values. For more information, see "Selecting
Multiple Columns for Matching" on page 390.
Note: Concatenating columns is not recommended for exact match
columns.
6. Click OK.
The Schema Manager adds the selected match column(s) to the Match
Columns list.
- 398 -
7. Click the Save button to save your changes.
- 399 -
Match rule sets allow you to execute different sets of match column rules at
different times. The match process uses only one match rule set per
execution. To match using a different match rule set, the match rule set must
be selected and the match process must be executed again.
Note: Only one match column rule in the match rule set needs to succeed in
order to declare a match between records.
You can configure any number of rule sets. When users want to run the Match
batch job, they select one rule set from the list of rule sets that have been
defined for the base object.
For more information about choosing match rule sets, see "Selecting a Match
Rule Set" on page 549.
In the Schema Manager, you designate one match rule set as the default.
Match rule sets allow you to accommodate different match column rule
requirements at different times. For example, you might use one match rule
set for an initial data load and a different match rule set for subsequent
incremental loads. Similarly, you might use one match rule set to process all
records, and another match rule set with a filter to process just a subset of
records (see "Filtering SQL" on page 403).
Before saving any changes to a match rule set (including any changes to
match rules in the match rule set), the Schema Manager analyzes the match
- 400 -
rule set and prompts you with a warning message if the match rule set has
any issues.
Note: This is only a warning message. You can choose to ignore the message
and save changes anyway.
Name
Search Levels
Used with fuzzy-match base objects only. When you configure a match rule
set, you define a search level that instructs Informatica MDM Hub on how
stringently and thoroughly to search for candidate matches.
The goal of the match process is to find the optimal number of matches for
your data:
• not too few (called undermatching), which misses relevant matches, or
- 401 -
• not too many (called overmatching), which generates too many matches,
including matches that are not relevant
For any name or address in a fuzzy match key, Informatica MDM Hub uses the
defined search level to generate different key ranges for the purpose of
determining which records are possible match candidates—and to which
records the match column rules will be applied.
The search level you choose should be determined by the size of your data
set, your time constraints, and how critical the matches are. Depending on
your circumstances and requirements, it is sometimes more appropriate to
undermatch, while at other times, it is more appropriate to overmatch.
Implementations dealing with relatively reliable and complete data can use
the Narrow level, while implementations dealing with less reliable data or with
more critical problems should use Exhaustive or Extreme.
- 402 -
records in a package based on match column and rule definitions. The
searchMatch request uses the columns in these records to generate match
columns that are used by the match server to find match candidates. For more
information about searchMatch, see the Informatica MDM Hub Services
Integration Framework Guide and the Informatica MDM Hub Javadoc.
By default, when an application calls the SIF searchMatch request, all possible
match columns are generated from the package or mapping records specified
in the request, and the match is performed by treating all columns with equal
weight. You can enable this option, however, to allow applications to specify
input match columns, in which case the searchMatch API ignores any columns
that were not passed as part of the request. You might use this feature if, for
example, you were using a custom population definition and wanted to call the
searchMatch API with a particular set of rules.
Enable Filtering
For example, if you had an Organization base object that contained multiple
types of organizations (customers, vendors, prospects, partners, and so on),
you could define different match rule sets that selectively processed only the
type of records you want to match: MatchAll (no filter), MatchCustomersOnly,
MatchVendorsOnly, and so on.
Filtering SQL
By default, when the Match batch job is run (see "Match Jobs" on page 547),
the match rule set processes all records. If the Enable Filtering check box (see
"Enable Filtering" on page 403) is selected (checked), you can specify a filter
condition to restrict processing to only those rules that meet the filter
condition. A filter is analogous to a WHERE clause in a SQL statement. The
filter expression can be any expression that is valid for the WHERE clause
syntax used in your database platform.
Note: The match rule set filter is applied to the base object records that are
selected for the match batch only (the records to match from)—not the
- 403 -
records in the match pool (the records to match to). For more information,
see "Flagging the Match Batch" on page 251.
All other, non-customer records would be ignored and not processed by the
Match job.
Match Rules
This area of the window displays a list of match column rules that have been
configured for the selected match rule set. For more information, see
"Configuring Match Column Rules for Match Rule Sets" on page 407.
- 404 -
Search Level Description
Match Rule Sets List of configured match rule sets.
Properties Properties for the selected match rule set.
4. Enter a unique, descriptive name for this new match rule set.
5. Click OK.
The Schema Manager adds the new match rule set to the list.
6. Configure the match rule set according to the instructions in the next
section, "Editing Match Rule Set Properties" on page 405.
- 405 -
• The following example shows the properties for an exact-match base
object.
4. Configure properties for this match rule set. For more information, see
"Match Rule Set Properties" on page 401.
5. Configure match columns for this match rule set according to the
instructions in "Configuring Match Column Rules for Match Rule Sets" on
page 407.
- 406 -
1. In the Schema Manager, display the Match Rule Sets tab in the
Match/Merge Setup Details dialog for the base object that you want to
configure. For more information, see "Navigating to the Match Rule Set
Tab" on page 404.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Select the match rule set that you want to rename.
- 407 -
information about the difference between match column rules and primary
key rules, see "Configuring Primary Key Match Rules" on page 434.
The properties for match column rules differ between exact match and fuzzy-
match base objects (see "Exact-match and Fuzzy-match Base Objects" on
page 247).
• For exact-match base objects, you can configure only exact column types.
• For fuzzy-match base objects, you can configure fuzzy or exact column
types. For more information, see "Match Rule Properties for Fuzzy-match
Base Objects Only" on page 409.
For each match column rule, decide whether matched records should be
automatically or manually consolidated. For more information, see
"Specifying Consolidation Options for Match Column Rules " on page 431 and
"Consolidating Records Automatically or Manually" on page 256.
- 408 -
Match Rule Properties for Fuzzy-match Base
Objects Only
This section describes match rule properties for fuzzy-match base objects.
These properties do not apply to exact-match base objects.
For fuzzy-match base objects, the match / search strategy defines the
strategy that Informatica MDM Hub uses for searching and matching in the
match rule. Select one of the following options:
Strategy Description
Option
Fuzzy Probabilistic match that takes into account spelling variations,
possible misspellings, and other differences that can make matching
records non-identical.
Exact Matches only records that are identical.
Certain configuration settings on the Match / Merge Setup tab apply to only
one type of column. In this document, such features are indicated with a
graphic that shows whether it applies to fuzzy-match columns only (as in the
following example), or exact-match columns only. No graphic means that the
feature applies to both.
- 409 -
Note: This match / search strategy is configured at the match rule level. For
more information about the match / search strategy configured at the base
object level (which determines whether it is a fuzzy-match base object or
exact-match base object), see "Match/Search Strategy" on page 370.
When specifying the match / search strategy for a fuzzy-match base object,
consider the implications of configuring the following types of match rules:
Type of Applies to
Match Rule
Fuzzy - Fuzzy Fuzzy and exact-match columns.
Search
Strategy
Exact - Exact Exact-match columns only. This option bypasses the fuzziness
Search of the base object and executes a simple exact match rule on a
Strategy fuzzy base object.
Filtered - Exact-match columns only. This option uses the fuzzy match
Fuzzy Search key as a filter, and then applies the exact match rule.
Strategy
Match Purpose
For fuzzy-match base objects, the match purpose defines the primary goal
behind a match rule. For example, if you're trying to identify matches for
people where address is an important part of determining whether two
records are for the same person, then you would choose the Match Purpose
called Resident.
For every match rule you define, you must choose the purpose of the rule
from a list of predefined match purposes provided by Informatica. Each match
purpose contains knowledge about how best to compare two records to
achieve the purpose of the match. Informatica MDM Hub uses the selected
match purpose as a basis for applying the match rules to determine matched
records. The behavior of the rules is dependent on the selected purpose. The
list of available match purposes depends on the population used, as described
in "Fuzzy Population" on page 370.
- 410 -
Two rules with all attributes identical (except for the purpose) will return
different sets of matches because of the different purpose.
Name Formats
Informatica MDM Hub match has the concept of a default name format which
tells it where to expect the last name. The options are:
• Left—last name is at the start of the full name, for example Smith Jim
• Right—last name is at the end of the full name, for example, Jim Smith
The name format used by Informatica MDM Hub depends on the purpose that
you're using. If you are using Organization, then the default is Last name,
First name, Middle name. If using Person/Resident then the default is First
Middle Last.
Bear this in mind when formatting data for matching. It might not make a big
difference, but there are edge cases where it helps, particularly for names
that do not fall within the selected population.
- 411 -
Match Purpose Settings
Match Description
Purpose
Person_ This purpose is for matches intended to identify a person by
Name name. This purpose is best suited to online searches when a
name-only lookup is required and a human is available to make
the choice. Matching in batch typically requires other attributes
in addition to name to make match decisions. Use this purpose
only when the rule does not contain address fields. This purpose
will allow matches between people with an address and those
without an address. If the rules contain address fields, use the
Resident purpose instead.
This purpose uses the following fields:
• Person_Name (Required)
• Address_Part1
• Address_Part2
• Postal_Area
• Telephone_Number
• ID
• Date
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, use Postal_Area as a repeat value in the Address_Part2
field.
Individual This purpose is intended to identify a specific individual by name
and with either the same ID number or date of birth attributes.
Since this purpose requires additional information, it is typically
used after a search by Person_Name.
This purpose uses the following fields:
• Person_Name (Required)
• ID-Either ID or Date are required (Using both is acceptable.)
• Date
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
Resident Intended to identify a person at an address. This purpose is
typically used after a search by either Person_Name or Address_
Part1. Optional input fields help qualify or rank a match if more
information is available.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
This purpose uses the following fields:
• Person_Name (Required)
• Address_Part1 (Required)
• Address_Part2
• Postal_Area
• Telephone_Number
• ID
- 412 -
Match Description
Purpose
• Date
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
Household Designed to identify matches where individuals with the same or
similar family names share the same address.
This purpose is typically used after a search by Address_Part1.
(Note: it is not practical to search by Person_Name because
ultimately only one word from the Person_Name must match,
and a one-word search will not perform well in most situations).
Emphasis is placed on the Last Name, the major word of the
Person_Name field, so this is one of the few cases where word
order is important in the way the records are presented for
matching.
However, a reasonable score will be generated provided that a
match occurs between the major word in one name and any
other word in the other name.
This purpose uses the following fields:
• Person_Name (Required)
• Address_Part1 (Required)
• Address_Part2
• Postal_Area
• Telephone_Number
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
- 413 -
Match Description
Purpose
Family Designed to identify matches where individuals with the same or
similar family names share the same address or the same
telephone number.
This purpose is typically used after a tiered search (multi-
search) by Address_Part1 and Telephone_Number. (Note: it is
not practical to search by Person_Name because ultimately only
one word from the Person_Name needs to match, and a one-
word search will not perform well in most situations).
Emphasis is placed on the Last Name, the major word of the
Person_Name field, so this is one of the few cases where word
order is important in the way the records are presented for
matching.
However, a reasonable score will be generated provided that a
match occurs between the major word in one name and any
other word in the other name.
This purpose uses the following fields:
• Person_Name (Required)
• Address_Part1 (Required)
• Telephone_Number (Required) (Score will be based on best
of Address_Part_1 and Telephone_Number)
• Address_Part2
• Postal_Area
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
- 414 -
Match Description
Purpose
Wide_ Designed to identify matches where the same address is shared
Household by individuals with the same family name or with the same
telephone number.
This purpose is typically used after a search by Address_Part1.
(Note: it is not practical to search by Person_Name because
ultimately only one word from the Person_Name needs to
match, and a one-word search will not perform well in most
situations).
Emphasis is placed on the last name, the major word of the
Person_Name field, so this is one of the few cases where word
order is important in the way the records are presented for
matching.
However, a reasonable score will be generated provided that a
match occurs between the major word in one name and any
other word in the other name.
This purpose uses the following fields:
• Address_Part1 (Required)
• Person_Name (Required)
• Telephone_Number (Required)
• Score will be based on best of Person_Name and Telephone_
Number
• Address_Part2
• Postal_Area
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
Address Designed to identify an address match. The address might be
postal, residential, delivery, descriptive, formal, or informal.
The only required field is Address_Part1. The fields Address_
Part2, Postal_Area, Telephone_Number, ID, Date, Attribute1 and
Attribute2 are available as optional input fields to further
differentiate an address. For example if the name of a City
and/or State is provided as Address_Part2, it will help
differentiate between a common street address [100 Main
Street] in different locations.
This purpose uses the following fields:
• Address_Part1 (Required)
• Address_Part2
• Postal_Area
• Telephone_Number
• ID
• Date
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2.
- 415 -
Match Description
Purpose
In that case, the Address_Part2 score used will be the higher of
the two scored fields.
Organization Designed to match organizations primarily by name. It is
targeted at online searches when a name-only lookup is required
and a human is available to make the choice. Matching in batch
typically requires other attributes in addition to name to make
match decisions. Use this purpose only when the rule does not
contain address fields. This purpose will allow matches between
organizations with an address and those without an address. If
the rules contain address fields, use the Division purpose.
This purpose uses the following fields:
• Organization_Name (Required)
• Address_Part1
• Address_Part2
• Postal_Area
• Telephone_Number
• ID
• Date
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required. Any optional
input fields you provide refine the ranking of matches.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
- 416 -
Match Description
Purpose
Division Designed to identify an Organization at an Address. It is typically
used after a search by Organization_Name or by Address_Part1,
or both.
It is in essence the same purpose as Organization, except that
Address_Part1 is a required field. Thus, this Purpose is designed
to match company X at an address of Y (or Z, etc., if multiple
addresses are supplied).
This purpose uses the following fields:
• Organization_Name (Required)
• Address_Part1 (Required)
• Address_Part2
• Postal_Area
• Telephone_Number
• ID
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
Contact Designed to identify a contact within an organization at a specific
location.
This Match purpose is typically used after a search by Person_
Name. However, either Organization_Name or Address_Part1
may be used as the search criteria.
This purpose uses the following fields:
• Person_Name (Required)
• Organization_Name (Required)
• Address_Part1 (Required)
• Address_Part2
• Postal_Area
• Telephone_Number
• ID
• Date
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
- 417 -
Match Description
Purpose
Corporate_ Designed to identify an Organization by its legal corporate
Entity name, including the legal endings such as INC, LTD, etc. It is
designed for applications that need to honor the differences
between such names as ABC TRADING INC and ABC TRADING
LTD.
This purpose is typically used after a search by Organization_
Name. It is in essence the same purpose as Organization, except
that tighter matching is performed and legal endings are not
treated as noise.
This purpose uses the following fields:
• Organization_Name (Required)
• Address_Part1
• Address_Part2
• Postal_Area
• Telephone_Number
• ID
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
To achieve a “best of” score between Address_Part2 and Postal_
Area, pass Postal_Area as a repeat value in the Address_Part2
field.
Wide_ Designed to loosely identify a contact within an organization—
Contact that is, without regard to actual location.
It is typically used after a search by Person_Name.
In addition to the required fields, ID, Attribute1 and Attribute2
may be optionally provided for matching to further qualify a
contact.
This purpose uses the following fields:
• Person_Name (Required)
• Organization_name (Required)
• ID
• Attribute1
• Attribute2
Unless otherwise indicated, fields are not required.
Fields Provided for general, non-specific use. It is designed in such a
way that there are no required fields. All field types are
available as optional input fields.
Match Levels
For fuzzy-match base objects, the match level determines how precise the
match is. You can specify one of the following match levels for a fuzzy-match
base object:
- 418 -
Match Levels
Level Description
Typical Appropriate for most matches.
Conservative Produces fewer matches than the Typical level. Some data that
actually matches may pass through the match process without
being flagged as a match. This situation is called
undermatching.
Loose Produces more matches than the Typical level. Loose matching
may produce a significant number of match candidates that are
not really matches. This situation is called overmatching. You
might choose to use this in a match rule for manual merges, to
make sure that other, tighter match rules have not missed any
potential matches.
Select the level based on your knowledge of the data to be matched: Typical,
Conservative (fewer matches), or Looser (more matches). When in doubt, use
Typical.
For fuzzy-match base objects, the accept limit is a number that determines
the acceptability of a match. This setting does the exact same thing as the
match level (see "Match Levels" on page 418), but to a more granular degree.
The accept limit is defined by Informatica within a population in accordance
with its match purpose. The Accept Limit Adjustment allows a coarse
adjustment to what is considered to be a match for this match rule.
• A positive adjustment results in more conservative matching.
• A negative adjustment results in looser matching.
For example, suppose that, for a given field and a given population, the accept
limit for a typical match level is 80, for a loose match level is 70, and for a
conservative match level is 90. If you specify a positive number (such as 3)
for the adjustment, then the accept level becomes slightly more conservative.
If you specify a negative number (such as -2), then the accept level becomes
looser.
- 419 -
Match Column Properties for Match Rules
This section describes the match column properties that you can configure for
match rules.
Match Subtype
For base objects containing different types of data, the match subtype option
allows you to apply match rules to specific types of data within the same base
object. You have the option to enable or disable match subtyping for exact-
match columns that have parent/child path components. Match subtype is
available only for:
• exact-match column types that are based on a non-root Path Component,
and
• match rules that have a fuzzy match / search strategy
To use match subtyping, for each match rule, specify one or more exact-
match column(s) that will serve as the “subtyping” column(s) to use. The
subtype indicator can be set for any of the exact-match columns regardless of
whether they are used for segment match or not. During the match process,
evaluation of the subtype column precedes evaluation of the other match
columns. Use match subtyping judiciously, because it can have a performance
impact on the match process.
- 420 -
Without Match Subtype, Parent ID 3 would match with 5 and 7. With Match
Subtype, however, Parent ID 3 will not match with 5 nor 7 because the
matching rows are distributed between different Address Types. Parent ID 5
and 7 will match with each other, however, because the matching rows all fall
within the 'Billing' Address Type.
Non-Equal Matching
NULL Matching
Note: Null Matching and Segment Matching are mutually exclusive. If one is
selected, then the other cannot be selected.
Use NULL matching to specify how the match process should behave when null
values match other null values. NULL matching applies only to exact-match
columns.
- 421 -
Property Description
Matches values are identical between records, then it is considered a
Non-NULL match.
Once null matching is configured, Build Match Groups will allow only a single
“Null to non NULL” match into any group, thereby reducing the possibility of
unwanted transitive matching. For more information, see "Build Match Groups
and Transitive Matches" on page 249.
Segment Matching
For exact-match columns only, you can use segment matching to limit match
rules to specific subsets of data. For example, you could define different
match rules for customers in different countries by using segment matching to
limit certain rules to specific country codes. Segment matching applies to both
exact-match and fuzzy-match base objects. For more information, see
"Configuring Segment Matching for a Column" on page 433.
If the Segment Matching check box is checked (selected), you can configure
two other options: Segment Matches All Data and Segment Match Values.
When unchecked (the default), Informatica MDM Hub will only match records
within the set of values defined in Segment Match Values. For example,
suppose a base object contained Leads, Partners, Customers, and Suppliers. If
Segment Match Values contained the values Leads and Partners, and Segment
Matches All Data were unchecked, then Informatica MDM Hub would only
match within records that contain Leads or Partners. All Customers and
Suppliers records will be ignored.
- 422 -
With Segment Matches All Data checked (selected), then Leads and Partners
would match with Customers and Suppliers, but Customers and Suppliers
would not match with each other.
For segment matching, specifies the list of segment values to use for segment
matching. You must specify one or more values (for a match column) that
defines the segment matching. For example, for a given match rule, suppose
you wanted to define segment matching by Gender. If you specified a segment
match value of M (for male), then, for that match rule, Informatica MDM Hub
searches for matches (based on the other match columns) only on male
records—and can only match to other male records, unless you also enabled
Segment Matches All Data.
- 423 -
text columns in the child base object. Matching on the match columns of a
child base object is called intertable matching.
• When using intertable match and creating match rules for the child table
(via a foreign key), you must include the foreign key from the parent table
in each match rule on the child. If you do not, when the child is merged,
the parent records would lose the child records that had previously
belonged to them.
For more information, see "Match Columns Depend on the Search Strategy" on
page 388.
Important: If you change your match rules after matching, you are prompted
to reset your matches. When you reset your matches, it deletes everything in
the match table and, in records where the consolidation indicator is 2, resets
the consolidation indicator to 4. For more information, see "About the
Consolidate Process" on page 255 and "Reset Match Table Jobs" on page 555.
- 424 -
1. In the Schema Manager, display the Match/Merge Setup Details dialog for
the base object that you want to configure. For more information, see
"Navigating to the Match/Merge Setup Details Dialog" on page 365.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click the Match Rule Sets tab. For more information, see "Navigating to
the Match Rule Set Tab" on page 404.
4. Select a match rule set in the list.
The Schema Manager displays the properties for the selected match rule
set.
5. In the Match Rules section of the screen, click the plus button .
The Schema Manager displays the Edit Match Rule dialog. This dialog
differs slightly between exact match and fuzzy-match base objects.
Exact-match Base Objects
- 425 -
Fuzzy-match Base Objects
6. For fuzzy-match base objects, configure the match rule properties at the
top of the dialog box. For more information, see "Match Rule Properties for
Fuzzy-match Base Objects Only" on page 409.
7. Configure the match column(s) for this match rule.
Only columns you have previously defined as match columns are shown.
• For exact-match base objects or match rules with an exact match /
search strategy, only exact column types are available.
• For fuzzy-match base objects, you can choose fuzzy or exact column
types.
- 426 -
For more information, see "Match Columns Depend on the Search
Strategy" on page 388.
b. Check (select) the check box next to any column that you want to
include.
c. Uncheck (clear) the check box next to any column that you want to
omit.
d. Click OK.
The Schema Manager displays the selected columns in the Match Columns
list.
8. Configure the match properties for each match column in the Match
Columns list. For more information, see:
• "Match Column Properties for Match Rules" on page 420
• "Configuring the Match Weight of a Column" on page 432
• "Configuring Segment Matching for a Column" on page 433
• "NULL Matching" on page 421
• "Match Subtype" on page 420
9. Click OK.
- 427 -
10. If this is an exact match, specify the match properties for this match rule.
For more information, see "Requirements for Exact-match Columns in
Match Column Rules" on page 423. Click OK.
5. In the Match Rules section of the screen, click the Edit button.
The Schema Manager displays the Edit Match Rule dialog. This dialog
differs slightly between exact match and fuzzy-match base objects. For
more information, see "Adding Match Column Rules" on page 424.
6. For fuzzy-match base objects, change the match rule properties at the top
of the dialog box, if you want. For more information, see "Match Rule
Properties for Fuzzy-match Base Objects Only" on page 409.
7. Configure the match column(s) for this match rule, if you want.
Only columns you have previously defined as match columns are shown.
• For exact-match base objects or match rules with an exact match /
search strategy, only exact column types are available.
• For fuzzy-match base objects, you can choose fuzzy or exact columns
types.
For more information, see "Match Columns Depend on the Search
Strategy" on page 388.
- 428 -
a. Click the Edit button next to the Match Columns list.
The Schema Manager displays the Add/Remove Match Columns dialog.
b. Check (select) the check box next to any column that you want to
include.
c. Uncheck (clear) the check box next to any column that you want to
omit.
d. Click OK.
The Schema Manager displays the selected columns in the Match Columns
list.
8. Change the match properties for any match column that you want to edit.
For more information, see:
• "Match Column Properties for Match Rules" on page 420
• "Configuring the Match Weight of a Column" on page 432
• "Configuring Segment Matching for a Column" on page 433
• "NULL Matching" on page 421
• "Match Subtype" on page 420
9. Click OK.
10. If this is an exact match, specify the match properties for this match rule.
For more information, see "Requirements for Exact-match Columns in
Match Column Rules" on page 423. Click OK.
- 429 -
1. In the Schema Manager, display the Match/Merge Setup Details dialog for
the exact-match base object that you want to configure. For more
information, see "Navigating to the Match/Merge Setup Details Dialog" on
page 365.
2. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
3. Click the Match Rule Sets tab. For more information, see "Navigating to
the Match Rule Set Tab" on page 404.
4. Select a match rule set in the list.
5. In the Match Rules section, select the match rule that you want to delete.
• Click the button to move the selected match rule down in the
execution sequence.
- 430 -
incongruences. For more information, see "Rule Set Evaluation" on page
400.
8. If you are prompted to confirm saving changes, click OK button to save
your changes.
Note: A base object cannot have more than 200 user-defined columns if it will
have match rules that are configured for automatic consolidation.
- 431 -
Configuring the Match Weight of a Column
For a fuzzy-match column, you can change its match weight in the Edit Match
Rule dialog box. For each column, Informatica MDM Hub assigns an internal
match weight, which is a number that indicates the importance of this column
(relative to other columns in the table) for matching. The match weight varies
according to the selected match purpose and population. For example, if the
match purpose is Person_Name, then Informatica MDM Hub, when evaluating
matches, views a data match in the name column with greater importance
than a data match in a different column (such as the address).
By adjusting the match weight of a column, you give added weight to, and
elevate the significance of, that column (relative to other columns) when
Informatica MDM Hub analyzes values for matches.
- 432 -
incongruences. For more information, see "Rule Set Evaluation" on page
400.
4. If you are prompted to confirm saving changes, click OK button to save
your changes.
• To add a value, click , type the value you want to add, and click
OK.
- 433 -
Before saving changes, the Schema Manager analyzes the match rule set
and prompts you with a message if the match rule set contains certain
incongruences. For more information, see "Rule Set Evaluation" on page
400.
7. If you are prompted to confirm saving changes, click OK button to save
your changes.
For example, two systems might use the same set of customer IDs. If both
systems provide information about customer XYZ123 using identical primary
key values, the two systems are certainly referring to the same customer and
the records should be automatically consolidated.
When you specify a primary key match, you simply specify which source
systems that have the same primary key values. You also check the Auto-
merge matching records check box to have Informatica MDM Hub
automatically consolidate matching records when a Merge or Link batch job is
run. For more information, see "Automerge Jobs" on page 534 and "Autolink
Jobs" on page 532.
- 434 -
The Primary Key Match Rules tab has the following columns.
Column Description
Key Two source systems for which this primary match key rule
Combination will be used for matching. These source systems must
already be defined in Informatica MDM Hub (see "Configuring
Source Systems" on page 264), and staging tables for this
base object must be associated with these source systems
(see "Configuring Staging Tables" on page 275).
Auto-Merge Specifies whether this primary key match rule results in
automatic or manual consolidation. For more information,
see "About the Consolidate Process" on page 255.
5. Check (select) the check box next to two source systems for which you
want to match records based on the primary key.
6. Check (select) the Auto-merge matching records check box if you are
certain that records with identical primary keys are matches.
You can change your choice for Auto-merge matching records later, if
you want.
7. Click OK.
The Schema Manager displays the new rule in the Primary Key Rule tab.
- 435 -
8. Click the Save button to save your changes.
The Schema Manager asks you whether you want to reset existing
matches.
9. Choose Yes. to delete all matches currently stored in the match table, if
you want.
4. Scroll to the primary key match rule that you want to edit.
- 436 -
5. Check or uncheck the Auto-merge matching records check box to
enable or disable auto-merging, respectively.
7. Choose Yes to delete all matches currently stored in the match table, if
you want.
4. Select the primary key match rule that you want to delete.
- 437 -
The Schema Manager asks you whether you want to reset existing
matches.
8. Choose Yes to delete all matches currently stored in your Match table, if
you want.
In the Match / Merge Setup Details pane of the Schema Manager, the Match
Keys Distribution tab allows you to investigate the distribution of match keys
in the match key table. This tool can assist you with identifying potential hot
spots in your data—high concentrations of match keys that could result in
overmatching—where the match process generates too many matches,
including matches that are not relevant. By knowing where hot spots occur in
your data, you can refine data cleansing and match rules to reduce hot spots
and generate an optimal distribution of match keys for use in the match
process. Ideally, you want to have a relatively even distribution across all
keys.
- 438 -
Components of the Match Keys Distribution Tab
The Match Key Distribution tab displays a histogram, match keys, and match
columns.
Histogram
The histogram displays the statistical distribution of match keys in the match
key table.
Axis Description
Key Starting character(s) of the match key. If no filter is applied (the
(X- default), this is the starting character of the match key. If a filter is
axis) applied, this is the starting sequence of characters in the match key,
beginning with the left-most character. For more information, see
"Filtering Match Keys" on page 440.
Count Number of match keys in the match key table that begins with the
(Y- starting character(s). Hotspots in the match key table show up as
axis) disproportionately tall spikes (high number of match keys), relative to
other characters in the histogram.
67
The Match Keys List on the Match Keys Distribution tab displays records in the
match key table. For each record, it displays cell data for the following
columns:
Column Description
Name
ROWID ROWID_OBJECT that uniquely identifies the record in the base
object that is associated with this match key.
KEY Generated match key. SSA_KEY column in the match key table.
Depending on the configured match rules and the nature of the data in a
record, a single record in the base object table can have multiple generated
match keys.
- 439 -
Paging Through Records in the Match Key Table
Use the following command buttons to navigate the records in the match key
table.
Button Description
Displays the first page of records in the match key table.
Match Columns
The Match Columns area on the Match Keys Distribution tab displays match
column data for the selected record in the match keys list. This is the SSA_
DATA column in the match key table. For each match column that is
configured for this base object (see "Configuring Match Columns" on page
387), it displays the column name and cell data.
The filter condition specifies the beginning string sequence for qualified match
keys, evaluated from left to right. For example, to view only match keys
beginning with the letter M, you would select M for the filter. To further restrict
match keys and view data for only the match keys that start with the letters MD
you would add the letter D to the filter. The longer the filter expression, the
more restrictive the display.
Setting a Filter
To set a filter:
• Click the vertical bar in the Histogram associated with the character you
want to add to the filter.
For example, suppose you started with the following default view in the
Histogram.
- 440 -
If you click the vertical bar above a character (such as the M character), the
Histogram refreshes and displays the distribution for all match keys beginning
with that character.
Note that the Match Keys List now displays only those match keys that meet
the filter condition.
Navigating Filters
- 441 -
Informatica MDM Hub provides a mechanism for selectively excluding records
from the match process. You might want to do this if, for example, your data
contained records that you wanted the match process to ignore.
To configure this feature, in the Schema Manager, you add a column named
EXCLUDE_FROM_MATCH to a base object. This column must be an integer type
with a default value of zero (0), as described in "Adding Columns" on page
108.
Once the table is populated and before running the Match job, to exclude a
record from matching, change its value in the EXCLUDE_FROM_MATCH column
to a one (1) in the Data Manager. When the Match job runs, only those records
with an EXCLUDE_FROM_MATCH value of zero (0) will be tokenized and
processed—all other records will be ignored. When the cell value is changed,
the DIRTY_IND for this record is set to 1 so that match keys will be
regenerated when the tokenize process is executed, as described in "Match
Tokens and Match Keys" on page 240.
- 442 -
Chapter 15: Configuring the
Consolidate Process
This chapter describes how to configure the consolidate process for your
Informatica MDM Hub implementation.
Chapter Contents
• "Before You Begin" on page 443
• "About Consolidation Settings" on page 443
• "Changing Consolidation Settings" on page 447
- 443 -
Note: If the Requeue on Parent Merge setting for a child base object is set to
2, in the event of a merging parent, the consolidation indicator will be set to 4
for the child record. For more information, see "Requeue On Parent Merge" on
page 91.
To specify an immutable source for a base object, click the drop-down list
next to Immutable Rowid Object and select a source system.
This list displays the source system(s) associated with this base object. Only
one source system can be designated an immutable source system. For more
information, see "Configuring Source Systems" on page 264.
- 444 -
Distinct Systems
A distinct system provides data that gets inserted into the base object without
being consolidated. Records from a distinct system will never match with
other records from the same system, but they can be matched to and from
other records in other systems (their CONSOLIDATION_IND is set to 4 on
load). You can specify distinct source systems and configure whether, for each
source system, records are consolidated automatically or manually.
The following example shows both options selected for the Billing system.
- 445 -
Auto Rules Only
For distinct systems only, you can enable this option to allow you to configure
what types of rules are executed for the associated distinct source system.
Check (select) this check box if you want Informatica MDM Hub to apply only
the automatic consolidation rules (not the manual consolidation rules) for this
distinct system. By default, this option is disabled (unchecked).
For child base objects, Informatica MDM Hub provides a cascade unmerge
feature that allows you to specify what happens if records in the parent base
object are unmerged. By default, this feature is disabled, so that unmerging
parent records does not unmerge associated child records. In the Unmerge
Child When Parent Unmerges portion near the bottom of the Merge Settings
tab, if you check (select) the Cascade Unmerge check box for a child base
object, when records in the parent object are unmerged, Informatica MDM
Hub also unmerges affected records in the child base object.
In the Unmerge Child When Parent Unmerges portion near the bottom of the
Merge Settings tab, the Schema Manager displays only those match-enabled
columns in the child base object that are configured with a foreign key. For
more information, see "Configuring Foreign-Key Relationships Between Base
Objects" on page 113.
In situations where a parent base object has multiple child base objects, you
can explicitly enable cascade unmerge for each child base object. Once
configured, when the parent base object is unmerged, then all affected
records in all associated child base objects are unmerged as well.
- 446 -
Considerations for Using Cascade Unmerge
Note: When cascade unmerge is enabled, the child record may not be
unmerged if a previous manual unmerge was done on the child base object.
When you enable the unmerge feature, it applies to the child table and the
child cross-reference table. Once enabled, if you then unmerge the parent
cross-reference, the original child cross-reference should be unmerged as
well. This feature has no impact on the parent—the feature operates on the
child tables to provide additional flexibility.
- 447 -
4. Change any of the following settings:
• "Immutable Rowid Object" on page 443
• "Distinct Systems" on page 445
• "Unmerge Child When Parent Unmerges (Cascade Unmerge)" on page
446
- 448 -
Chapter 16: Configuring the Publish
Process
This chapter describes how to configure the publish process for Informatica
MDM Hub data using message triggers and embedded message queues. For an
introduction, see "Publish Process" on page 260.
Chapter Contents
• "Before You Begin" on page 449
• "Configuration Steps for the Publish Process" on page 450
• "Starting the Message Queues Tool" on page 450
• "Configuring Global Message Queue Settings" on page 451
• "Configuring Message Queue Servers" on page 452
• "Configuring Outbound Message Queues" on page 454
• "Configuring Message Triggers" on page 456
• "JMS Message XML Reference" on page 464
- 449 -
• Read the introduction to the publish process in "Publish Process" on page
260
After you have configured message queues, you can review run-time activities
using the Audit Manager according to the instructions in "Auditing Message
Queues" on page 689.
- 450 -
Pane Description
Properties Shows the properties for the selected message queue.
pane
Monitoring Description
Setting
Receive Default is 0. Amount of time allowed to receive the
Timeout messages from the queue.
(milliseconds)
Receive Batch Default is 100. Maximum number of events processed and
Size placed in the message queue in a single pass.
Message Default is 300000. Amount of time to pause before polling
Check for inbound messages or processing outbound messages.
Interval The same value applies to both inbound and outbound
(milliseconds) message queues.
Out of sync If configured, periodically polls for ORS metadata and
check interval regenerates the XML message schema if subsequent
(milliseconds) changes have been made to design objects in the ORS. For
more information, see "Generating and Deploying ORS-
specific Schemas" on page 617.
By default, this feature is disabled—set to zero (0)—and is
available only if:
• Data Changes Monitoring is enabled.
• ORS-specific XML message schema has been generated
using the JMS Event Schema Manager.
Note: Make sure that this value is greater than or equal to
the Message Check Interval.
Click the button next to any property that you want to change.
- 451 -
Configuring Message Queue Servers
This section describes how to configure message queue servers for your
Informatica MDM Hub implementation.
WebSphere Properties
- 452 -
3. Right-click anywhere in the Navigation pane and choose Add Message
Queue Server.
The Message Queues tool displays the Add Message Queue Server dialog.
4. the Message Queues tool displays Specify the properties for this message
queue server. For more information, see "Message Queue Server
Properties" on page 452.
Click the button next to any property that you want to change.
- 453 -
Configuring Outbound Message Queues
This section describes how to configure outbound JMS message queues for
your Informatica MDM Hub implementation.
- 454 -
4. Specify the message queue properties. For more information, see
"Message Queue Properties" on page 454.
5. Click OK.
The Message Queues tool prompts you to choose the queue assignment.
Assignment Description
Leave Queue is currently unassigned and not in use. Select this
Unassigned option to use this queue as the outbound queue for
Informatica MDM Hub API responses, or to indicate that the
queue is currently unassigned and is not in use.
Use with Queue is currently assigned and is available for use by
Message message triggers that are defined in the Schema Manager
Queue according to the instructions in "Configuring Message
Triggers Triggers" on page 456.
Use Legacy Select (check) this option only if your Informatica MDM Hub
XML implementation requires that you use the legacy XML
message format (Informatica MDM Hub XU version) instead
of the current version of the XML message format. For more
information, see "Legacy JMS Message XML Reference" on
page 479.
- 455 -
4. Change the editable properties for this message queue. For more
information, see "Message Queue Properties" on page 454.
Click the button next to any property that you want to change.
5. Change the queue assignment, if you want.
You can use the same message queue for all triggers, or you can use a
different message queue for each trigger. In order for an action to trigger a
- 456 -
message trigger, the message queues must be configured, and a message
trigger must be defined for that base object and action.
The following types of events can cause a message trigger to be fired and a
message placed in the queue.
Events for Which Message Queue Rules Can Be Defined
Event Description
Add new • Add the data through the load process
data • Add the data through the Data Manager
• Add the data through the API verb using PUT or CLEANSE_PUT
(either through HTTP, SOAP, MQ, and so on)
Add new A new record with a PENDING state is created. Applies to state-
pending enabled base objects only.
data
Update • Update the data through the load process
existing • Update the data through the Data Manager
data
• Update the data through the API verb using PUT or CLEANSE_
PUT (either through HTTP, SOAP, MQ, and so on)
Note:
• If trust rules prevent the base object columns from being
updated, no message is generated.
• If one or more of the specified columns are updated, a single
message is generated. This single message includes data from
all of the cross-references in all output systems.
Update An existing record with a PENDING state is updated. Applies to
existing state-enabled base objects only. For more information, see "State
pending Management" on page 159
data
Update, • updating data when only the XREF has changed through the
only XREF load process
changed • updating data when only the XREF has changed through the
API using PUT or CLEANSE_PUT (either through HTTP, SOAP,
MQ, and so on)
Pending An XREF record with a PENDING state is updated. This includes
update, promotion of a record. Applies to state-enabled base objects only.
only XREF For more information, see "State Management" on page 159
changed
Merging • Manual Merge via Merge Manager
data • Merge via the API Verb (either though HTTP, SOAP, MQ etc.)
• Automatch and Merge
Merging Merging data when the base object has been updated
data, Base
object
updated
Unmerging • Unmerge the data through the Data Manager
data • Unmerge the data through the API verb using UNMERGE
(either through HTTP, SOAP, EJB etc.)
Accepting • Accepting a single record as unique via the Merge Manager
data as • Accepting multiple records as unique via the Merge Manager
unique
- 457 -
Event Description
• Having Accept as Unique turned on in the Base Object's Match
rules (this happens during the match/merge process)
Note: When a record is accepted as unique—either automatically
through a match rule or manually by a data steward—Informatica
MDM Hub generates a message with the record information,
including the cross-reference information for all output systems.
This message is placed in the queue.
Delete BO A base object record is soft deleted (state changed to DELETED).
data Applies to state-enabled base objects only. For more information,
see "State Management" on page 159
Delete An XREF record is soft deleted (state changed to DELETED).
XREF data Applies to state-enabled base objects only. For more information,
see "State Management" on page 159
Delete A base object record with a PENDING state is hard deleted.
pending Applies to state-enabled base objects only. For more information,
BO data see "State Management" on page 159
Delete An XREF record with a PENDING state is hard deleted. Applies to
pending state-enabled base objects only. For more information, see "State
XREF data Management" on page 159
No action Applies only to Activity Manager. Returned only by a cleanse_put
operation and only if delta detection is enabled. If delta detection
is not enabled, then an Update action type is returned.
Consider the following issues when setting up message triggers for your
implementation:
• If a message queue is used in any message trigger definition under a base
object in any Hub Store, the message queue displays the following
message: “The message queue is currently in use by message triggers.” In
this case, you cannot edit the properties of the message queue. Instead,
you must create another message queue to make the necessary changes.
• Message triggers apply to one base object only, and they fire only when a
specific action occurs directly on that base object. If you have two tables
that are in a parent-child relationship, then you need to explicitly define
message queues separately, for each table. Change detection is based on
specific changes to each base object (such as a load INSERT, load UPDATE,
MERGE, or PUT). Changes to a record of the parent table can fire a
message trigger for the parent record only. If changes in the parent record
affect one or more associated child records, then a message trigger for the
child table must be explicitly configured to fire when such an action occurs
in the child records.
- 458 -
2. Start the Schema Manager according to the instructions in "Starting the
Schema Manager" on page 81.
3. Acquire a write lock according to the instructions in "Acquiring a Write
Lock" on page 36.
4. Expand the base object that will be monitored, and select the Message
Trigger Setup node.
If no message triggers have been set up, then the Schema Tool displays an
empty screen.
- 459 -
6. Specify a name and description for the new message trigger.
7. Click Next.
The Add Message Trigger wizard prompts you to specify the messaging
package.
8. Select the package that will be used to build the message. For more
information, see "Configuring Packages" on page 151.
9. Click Next.
The Add Message Trigger wizard prompts you to specify the target
message queue.
10. Select the message queue to which the message will be written.
11. Click Next.
The Add Message Trigger wizard prompts you to specify the rules for this
message trigger.
- 460 -
12. Select the event type(s) for this message trigger.
For more information, see "Types of Events for Message Triggers" on page
457.
13. Configure the system properties for this message trigger:
Check Description
Box
Triggering System(s) that will trigger the action.
In For each message that is placed on a message queue due to the
Message trigger, the message includes the pkey_src_object value for
each cross-reference that it has in one of the 'In Message'
systems.
Note: You must select at least one Triggering system and one In Message
system.
For example, suppose your implementation had three source systems (A,
B, and C) and a base object record had cross-reference records for A and
B. Suppose the cross-reference in system A for this base object record
were updated. The following table shows possible message trigger
configurations and the resulting message:
- 461 -
In Message Systems Resulting Message
A Message with cross-reference for system A
B Message with cross-reference for system B
C No message – no cross-references from In Message
A& B Message with cross-reference for systems A and B
A& C Message with cross-reference for system A
B& C Message with cross-reference for system B
A& B&C Message with cross-reference for systems A and B
14. Identify the system to which the event applies, columns to listen to for
changes, and the package used to construct the message.
All events send the base object record—and all corresponding cross-
references that make up that record—to the message, based on the
specified package.
15. Click Next if you have selected an Update option. Otherwise click Finish.
16. If you have clicked the Update action, the Schema Manager prompts you
to select the columns to monitor for update actions.
- 462 -
3. Expand the base object that will be monitored, and select the Message
Trigger Setup node.
4. In the Message Triggers list, click the message trigger that you want to
configure.
The Schema Manager displays the settings for the selected message
trigger.
5. Change the settings you want. For more information, see "Adding Message
Triggers" on page 458 and "Types of Events for Message Triggers" on page
457.
Click the button next to editable property that you want to change.
- 463 -
JMS Message XML Reference
This section describes the structure of Informatica MDM Hub XML messages
and provides example messages.
Note: If your Informatica MDM Hub implementation requires that you use the
legacy XML message format (Informatica MDM Hub XU version) instead of the
current version of the XML message format (described in this section), see
"Legacy JMS Message XML Reference" on page 479 instead.
- 464 -
Field Description
<sourceSystemName> Name of the source system associated with this event.
<sourceKey> Value of the PKEY_SRC_OBJECT associated with this event.
<eventDate> Date/time when the event was generated.
<rowid> RowID of the base object record that was affected by the
event.
<xrefKey> Root node of a cross-reference record affected by this
event.
<systemName> System name of the cross-reference record affected by
this event.
<sourceKey> PKEY_SRC_OBJECT of the cross-reference record affected
by this event.
<packageName> Name of the secure package associated with this event.
<columnName> Each column in the package is represented by an element
in the XML file. Examples: rowidObject and
consolidationInd. Defined in the ORS-specific XSD that is
generated using the JMS Event Schema Manager tool. For
more information, see "Generating and Deploying ORS-
specific Schemas" on page 617.
<mergedRowid> List of ROWID_OBJECT values for the losing records in the
merge. This field is included in messages for Merge events
only.
Filtering Messages
You can use the custom JMS header named MessageType to filter incoming
messages based on the message type. The following message types are
indicated in the message header.
Message Description
Type
siperianEvent Event notification message.
< For Services Integration Framework (SIF) responses, the
serviceName response begins with the name of the SIF request, as in the
Return> following fragment of a response to a get request:
<getReturn>
<message>The GET was executed successfully - retrieved 1
records</message>
<recordKey>
<ROWID>2</ROWID>
</recordKey>
...
- 465 -
<?xml version="1.0" encoding="UTF-8"?>
<siperianEvent>
<eventMetadata>
<eventType>Accept as Unique</eventType>
<baseObjectUid>BASE_OBJECT.C_CONTACT</baseObjectUid>
<packageUid>PACKAGE.CONTACT_PKG</packageUid>
<orsId>localhost-mrm-CMX_ORS</orsId>
<triggerUid>MESSAGE_QUEUE_RULE.ContactUpdate</triggerUid>
<messageId>192</messageId>
<messageDate>2008-09-10T16:33:14.000-07:00</messageDate>
</eventMetadata>
<acceptAsUniqueEvent>
<sourceSystemName>Admin</sourceSystemName>
<sourceKey>SVR1.1T1</sourceKey>
<eventDate>2008-09-10T16:33:14.000-07:00</eventDate>
<rowid>2 </rowid>
<xrefKey>
<systemName>Admin</systemName>
<sourceKey>SVR1.1T1</sourceKey>
</xrefKey>
<contactPkg>
<rowidObject>2 </rowidObject>
<creator>admin</creator>
<createDate>2008-08-13T20:28:02.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-10T16:33:14.000-07:00</lastUpdateDate>
<consolidationInd>1</consolidationInd>
<lastRowidSystem>SYS0 </lastRowidSystem>
<dirtyInd>0</dirtyInd>
<firstName>Joey</firstName>
<lastName>Brown</lastName>
</contactPkg>
</acceptAsUniqueEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
AMRule Message
- 466 -
<eventDate>2008-09-19T11:43:42.979-07:00</eventDate>
<contactPkgAmEvent>
<amRuleUid>AM_RULE.RuleSet1|Rule1</amRuleUid>
<contactPkg>
<rowidObject>64 </rowidObject>
<creator>admin</creator>
<createDate>2008-09-08T16:24:35.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-18T16:26:45.000-07:00</lastUpdateDate>
<consolidationInd>2</consolidationInd>
<lastRowidSystem>SYS0 </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>Johnny</firstName>
<lastName>Brown</lastName>
<hubStateInd>1</hubStateInd>
</contactPkg>
<cContact>
<event>
<eventType>Update</eventType>
<system>Admin</system>
</event>
<event>
<eventType>Update XREF</eventType>
<system>Admin</system>
</event>
<xrefKey>
<systemName>CRM</systemName>
<sourceKey>PK1265</sourceKey>
</xrefKey>
<xrefKey>
<systemName>Admin</systemName>
<sourceKey>64</sourceKey>
</xrefKey>
</cContact>
</contactPkgAmEvent>
</amRuleEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
BoDelete Message
- 467 -
<boDeleteEvent>
<sourceSystemName>Admin</sourceSystemName>
<eventDate>2008-09-19T14:35:53.000-07:00</eventDate>
<rowid>107 </rowid>
<xrefKey>
<systemName>CRM</systemName>
</xrefKey>
<xrefKey>
<systemName>Admin</systemName>
</xrefKey>
<xrefKey>
<systemName>WEB</systemName>
</xrefKey>
<contactPkg>
<rowidObject>107 </rowidObject>
<creator>sifuser</creator>
<createDate>2008-09-19T14:35:28.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-19T14:35:53.000-07:00</lastUpdateDate>
<consolidationInd>4</consolidationInd>
<lastRowidSystem>CRM </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>John</firstName>
<lastName>Smith</lastName>
<hubStateInd>-1</hubStateInd>
</contactPkg>
</boDeleteEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
BoSetToDelete Message
<?xml version="1.0" encoding="UTF-8"?>
<siperianEvent>
<eventMetadata>
<eventType>BO set to Delete</eventType>
<baseObjectUid>BASE_OBJECT.C_CONTACT</baseObjectUid>
<packageUid>PACKAGE.CONTACT_PKG</packageUid>
<orsId>localhost-mrm-CMX_ORS</orsId>
<triggerUid>MESSAGE_QUEUE_RULE.ContactUpdate</triggerUid>
<messageId>319</messageId>
<messageDate>2008-09-19T14:21:03.000-07:00</messageDate>
</eventMetadata>
<boSetToDeleteEvent>
<sourceSystemName>Admin</sourceSystemName>
<eventDate>2008-09-19T14:21:03.000-07:00</eventDate>
<rowid>102 </rowid>
<xrefKey>
<systemName>CRM</systemName>
</xrefKey>
<xrefKey>
<systemName>Admin</systemName>
</xrefKey>
<xrefKey>
- 468 -
<systemName>WEB</systemName>
</xrefKey>
<contactPkg>
<rowidObject>102 </rowidObject>
<creator>admin</creator>
<createDate>2008-09-19T13:57:09.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-19T14:21:03.000-07:00</lastUpdateDate>
<consolidationInd>4</consolidationInd>
<lastRowidSystem>SYS0 </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<hubStateInd>-1</hubStateInd>
</contactPkg>
</boSetToDeleteEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Delete Message
- 469 -
<consolidationInd>4</consolidationInd>
<lastRowidSystem>CRM </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>John</firstName>
<lastName>Smith</lastName>
<hubStateInd>-1</hubStateInd>
</contactPkg>
</deleteEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Insert Message
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 470 -
Merge Message
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 471 -
<eventType>Merge Update</eventType>
<baseObjectUid>BASE_OBJECT.C_CONTACT</baseObjectUid>
<packageUid>PACKAGE.CONTACT_PKG</packageUid>
<orsId>localhost-mrm-CMX_ORS</orsId>
<triggerUid>MESSAGE_QUEUE_RULE.ContactUpdate</triggerUid>
<messageId>269</messageId>
<messageDate>2008-09-10T17:25:42.000-07:00</messageDate>
</eventMetadata>
<mergeUpdateEvent>
<sourceSystemName>CRM</sourceSystemName>
<sourceKey>P45678</sourceKey>
<eventDate>2008-09-10T17:25:42.000-07:00</eventDate>
<rowid>83 </rowid>
<xrefKey>
<systemName>CRM</systemName>
<sourceKey>P45678</sourceKey>
</xrefKey>
<mergedRowid>58 </mergedRowid>
<contactPkg>
<rowidObject>83 </rowidObject>
<creator>admin</creator>
<createDate>2008-09-10T16:44:56.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-10T17:25:42.000-07:00</lastUpdateDate>
<consolidationInd>1</consolidationInd>
<lastRowidSystem>CRM </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>Thomas</firstName>
<lastName>Jones</lastName>
</contactPkg>
</mergeUpdateEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
No Action Message
- 472 -
<xrefKey>
<systemName>CRM</systemName>
<sourceKey>P45678</sourceKey>
</xrefKey>
<xrefKey>
<systemName>CRM</systemName>
<sourceKey>P45678</sourceKey>
</xrefKey>
<xrefKey>
<systemName>CRM</systemName>
<sourceKey>P45678</sourceKey>
</xrefKey>
<contactPkg>
<rowidObject>83 </rowidObject>
<creator>admin</creator>
<createDate>2008-09-10T16:44:56.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-10T17:25:42.000-07:00</lastUpdateDate>
<consolidationInd>1</consolidationInd>
<lastRowidSystem>CRM </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>Thomas</firstName>
<lastName>Jones</lastName>
</contactPkg>
</noActionEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
PendingInsert Message
- 473 -
<creator>admin</creator>
<createDate>2008-09-19T13:57:09.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-19T13:57:09.000-07:00</lastUpdateDate>
<consolidationInd>4</consolidationInd>
<lastRowidSystem>SYS0 </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>John</firstName>
<lastName>Smith</lastName>
<hubStateInd>0</hubStateInd>
</contactPkg>
</pendingInsertEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
PendingUpdate Message
- 474 -
<hubStateInd>1</hubStateInd>
</contactPkg>
</pendingUpdateEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
PendingUpdateXref Message
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 475 -
Unmerge Message
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Update Message
- 476 -
<triggerUid>MESSAGE_QUEUE_RULE.ContactUpdate</triggerUid>
<messageId>120</messageId>
<messageDate>2008-09-08T16:05:13.000-07:00</messageDate>
</eventMetadata>
<updateEvent>
<sourceSystemName>CRM</sourceSystemName>
<sourceKey>PK12658</sourceKey>
<eventDate>2008-09-08T16:05:13.000-07:00</eventDate>
<rowid>66 </rowid>
<xrefKey>
<systemName>CRM</systemName>
<sourceKey>PK12658</sourceKey>
</xrefKey>
<contactPkg>
<rowidObject>66 </rowidObject>
<creator>admin</creator>
<createDate>2008-09-08T16:02:11.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-08T16:05:13.000-07:00</lastUpdateDate>
<consolidationInd>4</consolidationInd>
<lastRowidSystem>CRM </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>Joe</firstName>
<lastName>Black</lastName>
</contactPkg>
</updateEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 477 -
<rowidObject>66 </rowidObject>
<creator>admin</creator>
<createDate>2008-09-08T16:02:11.000-07:00</createDate>
<updatedBy>admin</updatedBy>
<lastUpdateDate>2008-09-08T16:05:13.000-07:00</lastUpdateDate>
<consolidationInd>4</consolidationInd>
<lastRowidSystem>CRM </lastRowidSystem>
<dirtyInd>1</dirtyInd>
<firstName>Joe</firstName>
<lastName>Black</lastName>
</contactPkg>
</updateXrefEvent>
</siperianEvent>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
XRefDelete Message
- 478 -
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
XRefSetToDelete Message
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 479 -
current version of the XML message format (described in "JMS Message XML
Reference" on page 464).
- 480 -
Example Messages for Legacy XML
This section provides listings of example messages.
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
BO Delete Message
- 481 -
<CONTROLAREA>
<ACTION>BO Delete</ACTION>
<MESSAGE_DATE>2008-09-19 14:35:53.0</MESSAGE_DATE>
<TABLE_NAME>C_CONTACT</TABLE_NAME>
<PACKAGE>CONTACT_PKG</PACKAGE>
<RULE_NAME>ContactUpdateLegacy</RULE_NAME>
<RULE_ID>SVR1.28D</RULE_ID>
<ROWID_OBJECT>107 </ROWID_OBJECT>
<DATABASE>localhost-mrm-CMX_ORS</DATABASE>
<XREFS>
<XREF>
<SYSTEM>CRM</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
<XREF>
<SYSTEM>Admin</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
<XREF>
<SYSTEM>WEB</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
</XREFS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_OBJECT>107 </ROWID_OBJECT>
<CREATOR>sifuser</CREATOR>
<CREATE_DATE>19 Sep 2008 14:35:28</CREATE_DATE>
<UPDATED_BY>admin</UPDATED_BY>
<LAST_UPDATE_DATE>19 Sep 2008 14:35:53</LAST_UPDATE_DATE>
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<DELETED_IND />
<DELETED_BY />
<DELETED_DATE />
<LAST_ROWID_SYSTEM>CRM </LAST_ROWID_SYSTEM>
<DIRTY_IND>1</DIRTY_IND>
<INTERACTION_ID />
<FIRST_NAME>John</FIRST_NAME>
<LAST_NAME>Smith</LAST_NAME>
<HUB_STATE_IND>-1</HUB_STATE_IND>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
BO set to Delete
- 482 -
<MESSAGE_DATE>2008-09-19 14:21:03.0</MESSAGE_DATE>
<TABLE_NAME>C_CONTACT</TABLE_NAME>
<PACKAGE>CONTACT_PKG</PACKAGE>
<RULE_NAME>ContactUpdateLegacy</RULE_NAME>
<RULE_ID>SVR1.28D</RULE_ID>
<ROWID_OBJECT>102 </ROWID_OBJECT>
<DATABASE>localhost-mrm-CMX_ORS</DATABASE>
<XREFS>
<XREF>
<SYSTEM>CRM</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
<XREF>
<SYSTEM>Admin</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
<XREF>
<SYSTEM>WEB</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
</XREFS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_OBJECT>102 </ROWID_OBJECT>
<CREATOR>admin</CREATOR>
<CREATE_DATE>19 Sep 2008 13:57:09</CREATE_DATE>
<UPDATED_BY>admin</UPDATED_BY>
<LAST_UPDATE_DATE>19 Sep 2008 14:21:03</LAST_UPDATE_DATE>
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<DELETED_IND />
<DELETED_BY />
<DELETED_DATE />
<LAST_ROWID_SYSTEM>SYS0 </LAST_ROWID_SYSTEM>
<DIRTY_IND>1</DIRTY_IND>
<INTERACTION_ID />
<FIRST_NAME />
<LAST_NAME />
<HUB_STATE_IND>-1</HUB_STATE_IND>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Delete Message
- 483 -
<PACKAGE>CONTACT_PKG</PACKAGE>
<RULE_NAME>ContactUpdateLegacy</RULE_NAME>
<RULE_ID>SVR1.28D</RULE_ID>
<ROWID_OBJECT>107 </ROWID_OBJECT>
<DATABASE>localhost-mrm-CMX_ORS</DATABASE>
<XREFS>
<XREF>
<SYSTEM>CRM</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
<XREF>
<SYSTEM>Admin</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
<XREF>
<SYSTEM>WEB</SYSTEM>
<PKEY_SRC_OBJECT />
</XREF>
</XREFS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_OBJECT>107 </ROWID_OBJECT>
<CREATOR>sifuser</CREATOR>
<CREATE_DATE>19 Sep 2008 14:35:28</CREATE_DATE>
<UPDATED_BY>admin</UPDATED_BY>
<LAST_UPDATE_DATE>19 Sep 2008 14:35:53</LAST_UPDATE_DATE>
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<DELETED_IND />
<DELETED_BY />
<DELETED_DATE />
<LAST_ROWID_SYSTEM>CRM </LAST_ROWID_SYSTEM>
<DIRTY_IND>1</DIRTY_IND>
<INTERACTION_ID />
<FIRST_NAME>John</FIRST_NAME>
<LAST_NAME>Smith</LAST_NAME>
<HUB_STATE_IND>-1</HUB_STATE_IND>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Insert Message
- 484 -
<XREFS>
<XREF>
<SYSTEM>CRM</SYSTEM>
<PKEY_SRC_OBJECT>49 </PKEY_SRC_OBJECT>
</XREF>
</XREFS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_OBJECT>33 </ROWID_OBJECT>
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<FIRST_NAME>James</FIRST_NAME>
<MIDDLE_NAME>Neville</MIDDLE_NAME>
<LAST_NAME>Darwent</LAST_NAME>
<SUFFIX>Unknown</SUFFIX>
<GENDER>M </GENDER>
<BIRTH_DATE>1938-06-22</BIRTH_DATE>
<SALUTATION>Mr</SALUTATION>
<SSN_TAX_NUMBER>216275400</SSN_TAX_NUMBER>
<FULL_NAME>James Darwent,Stony Brook Ny</FULL_NAME>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Merge Message
- 485 -
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<FIRST_NAME>Jimmy</FIRST_NAME>
<MIDDLE_NAME>Neville</MIDDLE_NAME>
<LAST_NAME>Darwent</LAST_NAME>
<SUFFIX>Jr</SUFFIX>
<GENDER>M </GENDER>
<BIRTH_DATE>1938-06-22</BIRTH_DATE>
<SALUTATION>Mr</SALUTATION>
<SSN_TAX_NUMBER>659483774</SSN_TAX_NUMBER>
<FULL_NAME>Jimmy Darwent, Stony Brook Ny</FULL_NAME>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 486 -
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 487 -
Pending Update Message
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 488 -
<SIP_EVENT>
<CONTROLAREA>
<ACTION>Pending Update XREF</ACTION>
<MESSAGE_DATE>2008-09-19 14:01:36.0</MESSAGE_DATE>
<TABLE_NAME>C_CONTACT</TABLE_NAME>
<PACKAGE>CONTACT_ADDRESS_PKG</PACKAGE>
<RULE_NAME>ContactAM</RULE_NAME>
<RULE_ID>SVR1.1VU</RULE_ID>
<ROWID_OBJECT>102 </ROWID_OBJECT>
<DATABASE>localhost-mrm-CMX_ORS</DATABASE>
<XREFS>
<XREF>
<SYSTEM>CRM</SYSTEM>
<PKEY_SRC_OBJECT>CPK125</PKEY_SRC_OBJECT>
</XREF>
<XREF>
<SYSTEM>Admin</SYSTEM>
<PKEY_SRC_OBJECT>SVR1.2V3</PKEY_SRC_OBJECT>
</XREF>
</XREFS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_CONTACT>102 </ROWID_CONTACT>
<CREATOR>admin</CREATOR>
<CREATE_DATE>19 Sep 2008 13:57:09</CREATE_DATE>
<UPDATED_BY>sifuser</UPDATED_BY>
<LAST_UPDATE_DATE>19 Sep 2008 14:01:36</LAST_UPDATE_DATE>
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<DELETED_IND />
<DELETED_BY />
<DELETED_DATE />
<LAST_ROWID_SYSTEM>CRM </LAST_ROWID_SYSTEM>
<DIRTY_IND>1</DIRTY_IND>
<INTERACTION_ID />
<FIRST_NAME>John</FIRST_NAME>
<LAST_NAME>Smith</LAST_NAME>
<HUB_STATE_IND>1</HUB_STATE_IND>
<CITY />
<STATE />
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Update Message
- 489 -
<RULE_NAME>CustomerRule1</RULE_NAME>
<RULE_ID>SVR1.8EO</RULE_ID>
<ROWID_OBJECT>74 </ROWID_OBJECT>
<SOURCE_XREF>
<SYSTEM>Admin</SYSTEM>
<PKEY_SRC_OBJECT>196 </PKEY_SRC_OBJECT>
</SOURCE_XREF>
<XREFS>
<XREF>
<SYSTEM>CRM</SYSTEM>
<PKEY_SRC_OBJECT>196 </PKEY_SRC_OBJECT>
</XREF>
<XREF>
<SYSTEM>SFA</SYSTEM>
<PKEY_SRC_OBJECT>49 </PKEY_SRC_OBJECT>
</XREF>
<XREF>
<SYSTEM>Admin</SYSTEM>
<PKEY_SRC_OBJECT>74 </PKEY_SRC_OBJECT>
</XREF>
</XREFS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_OBJECT>74 </ROWID_OBJECT>
<CONSOLIDATION_IND>1</CONSOLIDATION_IND>
<FIRST_NAME>Jimmy</FIRST_NAME>
<MIDDLE_NAME>Neville</MIDDLE_NAME>
<LAST_NAME>Darwent</LAST_NAME>
<SUFFIX>Jr</SUFFIX>
<GENDER>M </GENDER>
<BIRTH_DATE>1938-06-22</BIRTH_DATE>
<SALUTATION>Mr</SALUTATION>
<SSN_TAX_NUMBER>659483773</SSN_TAX_NUMBER>
<FULL_NAME>Jimmy Darwent, Stony Brook Ny</FULL_NAME>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 490 -
<PKEY_SRC_OBJECT>196 </PKEY_SRC_OBJECT>
</SOURCE_XREF>
<XREFS>
<XREF>
<SYSTEM>CRM</SYSTEM>
<PKEY_SRC_OBJECT>196 </PKEY_SRC_OBJECT>
</XREF>
<XREF>
<SYSTEM>SFA</SYSTEM>
<PKEY_SRC_OBJECT>49 </PKEY_SRC_OBJECT>
</XREF>
<XREF>
<SYSTEM>Admin</SYSTEM>
<PKEY_SRC_OBJECT>74 </PKEY_SRC_OBJECT>
</XREF>
</XREFS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_OBJECT>74 </ROWID_OBJECT>
<CONSOLIDATION_IND>1</CONSOLIDATION_IND>
<FIRST_NAME>Jimmy</FIRST_NAME>
<MIDDLE_NAME>Neville</MIDDLE_NAME>
<LAST_NAME>Darwent</LAST_NAME>
<SUFFIX>Jr</SUFFIX>
<GENDER>M </GENDER>
<BIRTH_DATE>1938-06-22</BIRTH_DATE>
<SALUTATION>Mr</SALUTATION>
<SSN_TAX_NUMBER>659483773</SSN_TAX_NUMBER>
<FULL_NAME>Jimmy Darwent, Stony Brook Ny</FULL_NAME>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
Unmerge Message
- 491 -
</XREFS>
<MERGED_OBJECTS>
<ROWID_OBJECT>0</ROWID_OBJECT>
</MERGED_OBJECTS>
</CONTROLAREA>
<DATAAREA>
<DATA>
<ROWID_OBJECT>10</ROWID_OBJECT>
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<LAST_ROWID_SYSTEM>SVR1.7NK</LAST_ROWID_SYSTEM>
<DIRTY_IND>1</DIRTY_IND>
<INTERACTION_ID />
<CONSUMER_ID>8</CONSUMER_ID>
<FIRST_NAME>THOMAS</FIRST_NAME>
<MIDDLE_NAME>L</MIDDLE_NAME>
<LAST_NAME>KIDD</LAST_NAME>
<SUFFIX />
<TELEPHONE>2178952323</TELEPHONE>
<GENDER>M</GENDER>
<DOB>1940</DOB>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 492 -
<CONSOLIDATION_IND>4</CONSOLIDATION_IND>
<DELETED_IND />
<DELETED_BY />
<DELETED_DATE />
<LAST_ROWID_SYSTEM>CRM </LAST_ROWID_SYSTEM>
<DIRTY_IND>1</DIRTY_IND>
<INTERACTION_ID />
<FIRST_NAME />
<LAST_NAME />
<HUB_STATE_IND>1</HUB_STATE_IND>
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 493 -
</DATA>
</DATAAREA>
</SIP_EVENT>
Your messages will not look exactly like this. The data will reflect your data,
and the fields will reflect your packages.
- 494 -
Part 4: Executing Informatica MDM Hub Processes
Contents
• "Using Batch Jobs " on page 496
• "Writing Custom Scripts to Execute Batch Jobs " on page 559
- 495 -
Chapter 17: Using Batch Jobs
This chapter describes how to configure and execute Informatica MDM Hub
batch jobs using the Batch Viewer and Batch Group tools in the Hub Console.
For more information about creating batch jobs using job execution scripts,
see "Writing Custom Scripts to Execute Batch Jobs " on page 559.
Chapter Contents
• "Before You Begin" on page 496
• "About Informatica MDM Hub Batch Jobs" on page 496
• "Running Batch Jobs Using the Batch Viewer Tool" on page 501
• "Running Batch Jobs Using the Batch Group Tool" on page 512
• "Batch Jobs Reference" on page 530
- 496 -
• Batch Group tool—Execute batch jobs in a group. The Batch Group
tool allows you to configure the execution sequence for batch jobs and
to execute batch jobs in parallel. For more information, see "Running
Batch Jobs Using the Batch Group Tool" on page 512.
• Stored procedures—Execute public Informatica MDM Hub processes
(batch jobs and batch groups) through stored procedures using any job
scheduling software (such as Tivoli, CA Unicenter, and so on). For more
information, see "About Executing Informatica MDM Hub Batch Jobs" on
page 559. You can also create and run stored procedures using the SIF API
(using Java, SOAP, or HTTP/XML). For more information, see the
Informatica MDM Hub Services Integration Framework Guide.
• Services Integration Framework (SIF) requests—Applications can
invoke the SIF ExecuteBatchGroupRequest request to execute batch groups
directly. For more information, see the Informatica MDM Hub Services
Integration Framework Guide.
- 497 -
organization should consider the best practice of developing an administration
or operations plan that specifies which batch processes and dependencies
should be completed before running batch jobs.
One of the tasks Informatica MDM Hub batch jobs perform is to move data
from landing tables to the appropriate target location in Informatica MDM Hub.
Therefore, before you run Informatica MDM Hub batch jobs, you must first
have your source systems or an ETL tool write data into the landing tables.
The landing tables are Informatica MDM Hub’s interface for batch loads. You
deliver the data to the landing tables, and Informatica MDM Hub batch
procedures manipulate the data and copy it to the appropriate location(s). For
more information, see the description of the Informatica MDM Hub data
management process in the Informatica MDM Hub Overview.
The general rule of thumb is that all parent tables (tables that other tables
reference) must be loaded first.
If two tables have a foreign key relationship between them, you must load the
table that is being referenced gets loaded first, and the table doing the
referencing gets loaded second. The following foreign key relationships can
exist in Informatica MDM Hub: from one base object (child with foreign key)
to another base object (parent with primary key).
In most cases, you will schedule these jobs to run on a regular basis.
- 498 -
The schema is fundamental to all your Informatica MDM Hub tasks.
Without a schema, your batch jobs have nothing to do. For more
information about defining the schema, see "About the Schema" on page
73.
• Define mappings before executing Stage jobs.
Mappings define the transformations performed in Stage jobs. If you have
no mappings defined, then the Stage job will not perform any
transformations in the staging process. For more information about
mappings, see "Mapping Columns Between Landing and Staging Tables" on
page 286.
• Define match rules before executing Match jobs.
If you have no match rules, then the Match job will produce no matches.
For more information, see "Configuring Primary Key Match Rules" on page
434.
• Before running production jobs:
• Run tests with small data sets.
• Run tests of your cleanse engine and other components to determine
whether each component is working as expected.
• After testing each of the components separately, test the integrated
system in its entirety to determine whether the overall system is
working as expected.
When you configure your Hub Store, the following types of batch jobs are
automatically created:
• "Auto Match and Merge Jobs" on page 532
• "Autolink Jobs" on page 532
• "Automerge Jobs" on page 534
• "BVT Snapshot Jobs" on page 535
• "External Match Jobs" on page 535
• "Generate Match Tokens Jobs" on page 540
• "Load Jobs" on page 542
• "Manual Link Jobs" on page 545
- 499 -
• "Manual Merge Jobs" on page 545
• "Manual Unlink Jobs" on page 546
• "Manual Unmerge Jobs" on page 546
• "Match Jobs" on page 547
• "Match Analyze Jobs" on page 550
• "Migrate Link Style To Merge Style Jobs" on page 552
• "Promote Jobs" on page 552
• "Reset Links Jobs" on page 555
• "Stage Jobs" on page 556
The following batch jobs are created when you make changes to the match
and merge setup, set properties, or enable trust settings after initial loads:
• "Accept Non-Matched Records As Unique " on page 532
• "Key Match Jobs" on page 541
• "Reset Links Jobs" on page 555
• "Reset Match Table Jobs" on page 555
• "Revalidate Jobs" on page 556 (if you enable validation for a column)
• "Synchronize Jobs" on page 557
- 500 -
Running Batch Jobs Using the Batch Viewer
Tool
This section describes how to use the Batch Viewer tool in the Hub Console to
run batch jobs individually. To run batch jobs in a group, see "Running Batch
Jobs Using the Batch Group Tool" on page 512.
Note: The Batch Viewer does not provide automated scheduling. For more
information about how to create custom scripts to execute batch jobs and
batch groups, see "About Executing Informatica MDM Hub Batch Jobs" on page
559.
- 501 -
Group By Option Description
Table Displays items in the hierarchy at the following levels:
• top level: tables
• second level: procedure type
• third level: batch job
• fourth level: date / timestamp
Date Displays items in the hierarchy at the following levels:
• top level: date / timestamp
• second level: batch jobs by date/timestamp
Procedure Type Displays items in the hierarchy at the following levels:
• top level: procedure type
• second level: batch job
• third level: date / timestamp
- 502 -
2. Expand the tree to display the batch job that you want to run, and then
click it to select it.
The Batch Viewer displays a screen for the selected batch job with properties
and command buttons.
Field Description
Identity Identification information for this batch job. Stored in the C_
REPOS_TABLE_OBJECT_V table
Name Type code for this batch job. For example, s have the
CMXLD.LOAD_MASTER type code. Stored in the OBJECT_NAME
column of the C_REPOS_TABLE_OBJECT_V table.
Description Description for this batch job in the format:
- 503 -
Field Description
JobName for | from BaseObjectName
Examples:
• Load from Consumer_Credit_Stg
• Match for Address
This description is stored in the OBJECT_DESC column of the C_
REPOS_TABLE_OBJECT_V table.
Status Status information for this batch job
Current Current status of the job. Examples:
Status • Executing
• Incomplete
• Completed
• Not Executing
• <Batch Job> Successful
• Description of failure
Certain types of batch jobs have additional fields that you can configure before
running the batch job.
Field Only For Description
Re- Generate Controls the scope of match tokens generation: tokenizes
generate Match the entire base object (checked) or tokenizes only those
All Token records that are flagged in the base object as requiring re-
Match Jobs tokenization (un-checked). For more information, see
Tokens "Regenerating All Match Tokens" on page 540.
Force Load If selected, the Load job forces a refresh and loads records
Update Jobs from the staging table to the base object regardless of
whether the records have already been loaded. For more
information, see "Forcing Updates in Load Jobs" on page
544.
Match Match Enables you to choose which match rule set to use for this
Set Jobs match job. For more information, see "Selecting a Match
Rule Set" on page 549.
After you have selected a batch job, you can click the following command
buttons.
.
Button Description
Executes the selected batch job.
- 504 -
Executing a Batch Job
Important: You must have the application server running for the duration of
an executing batch job.
To execute batch jobs in other ways, see "Ways to Execute Batch Jobs" on
page 496.
While a batch job is running, you can click Refresh Status to check if the
status has changed.
In very rare circumstances, you might want to change the status of a running
job by clicking Set Status to Incomplete and execute the job again. Only do
this if the batch job has stopped executing (due to an error, such as a server
reboot or crash) but Informatica MDM Hub has not detected that the job has
stopped due to a job application lock in the metadata. You will know this is a
problem if the current status is Executing but the database, application
server, and logs show no activity. If this occurs, click this button to clear the
job application lock so that you can run the batch job again; otherwise, you
will not be able to execute the batch job. Setting the status to Incomplete just
updates the status of the batch job—it does not abort the job.
- 505 -
Viewing Job Execution Logs
Informatica MDM Hub creates a job execution log each time that it executes a
batch job.
Each job execution log entry has one of the following status values:
Icon Description
Batch job is currently running.
- 506 -
Job Execution Log Entry Properties
For each job execution log entry, the Batch Viewer displays the following
information:
Field Description
Identity Identification information for this batch job. Stored in the C_REPOS_TABLE_OBJECT_V
table
Name Name of this job execution log. Date / time when the batch job
started.
Description Description for this batch job in the format:
JobName for / from BaseObjectName
Examples:
• Load from Consumer_Credit_Stg
• Match for Address
Source One of the following:
system • source system of the processed data
• Admin
Source Source table of the processed data.
table
Status Status information for this batch job
Current Current status of this batch job. If an error occurred, displays
Status information about the error. For more information, see "Job
Execution Status" on page 506.
Metrics Metrics for this batch job
[Various] Statistics collected during the execution of the batch job (if
applicable). For more information, see:
• "About Batch Job Metrics" on page 508
• "Auto Match and Merge Metrics" on page 533
• "Automerge Metrics" on page 534
• "Load Job Metrics" on page 544
- 507 -
Field Description
• "Match Job Metrics" on page 549
• "Match Analyze Job Metrics" on page 551
• "Stage Job Metrics" on page 557
• "Promote Job Metrics" on page 554
Time Timestamp for this batch job
Start Date / time when this batch job started.
Stop Date / time when this batch job ended.
Elapsed Elapsed time for the execution of this batch job.
time
- 508 -
Metric Name Description
Setup configuration. For more information, see "Accept All
Unmatched Rows as Unique" on page 369.
Queued for Number of records that were queued for automerge by a Match
automerge job that was executed by the Auto Match and Merge job. For
more information, see "Automerge Jobs" on page 534.
Queued for Number of records that were queued for manual merge. Use the
manual Merge Manager in the Hub Console to process these records. For
merge more information, see the Informatica MDM Hub Data Steward
Guide.
Backfill trust
records
Missing Number of source records that were missing lookup information
lookup / or had invalid rowid_object records.
Invalid
rowid_
object
records
Records Number of records placed on Hold status.
moved to
Hold status
Records Number of records to be matched.
analyzed (to
be matched)
Match Number of match comparisons.
comparisons
Total Number of cleansed records.
cleansed
records
Total Number of records placed in landing table.
landing
records
Invalid Number of records with invalid rowid_object.
supplied
rowid_
object
records
Auto-linked Number of auto-linked records.
records
BVT Snapshot of BVT.
snapshot
Duplicate Number of duplicate matched records.
matched
records
Links Number of links removed.
removed
Revalidated Number of records revalidated.
records
Base object Number of base object records reset to “new” status.
records
reset to New
status
- 509 -
Metric Name Description
Links Number of links converted to matches.
converted to
matches
Auto- Number of auto-promoted records.
promoted
records
Deleted Number of XREF records deleted.
XREF
records
Deleted Number of records deleted.
record
Invalid Number of invalid records.
records
Not Number of active records not promoted.
promoted
active
records
Not Number of protected records not promoted.
promoted
protected
records
Deleted BO Number of base object records deleted.
records
For Stage jobs or s only, if the batch job resulted in records being written to
the rejects table, then the job execution log displays a View Rejects button.
- 510 -
To view the rejected records and the reason why each was rejected:
1. Click the View Rejects button.
The Batch Viewer displays a table of rejected records.
2. Click Close.
To copy the current status of a batch to the Windows Clipboard (to paste into a
document or e-mail, for example):
• Click the button in the top right hand corner of the job properties page.
- 511 -
Clearing the Job Execution History
After running batch jobs over time, the list of executed jobs can become very
large. You should periodically remove the extraneous job execution logs from
this list.
Note: The actual procedure steps to clear job history will be slightly different
depending on the view (By Table, By Date, or By Procedure Type); the
following procedure assumes you are using the By Table view.
The Batch Viewer does not provide automated scheduling. For more
information about how to create custom scripts to execute batch jobs and
batch groups, see "Writing Custom Scripts to Execute Batch Jobs " on page 559
- 512 -
batch job in a batch group can be executed sequentially or in parallel with
other jobs. You use the Batch Group tool to configure and run batch groups.
For more information about batch jobs, see "Batch Jobs Reference" on page
530.
For more information about developing custom batch jobs and batch groups
that can be made available in the Batch Group tool, see "Developing Custom
Stored Procedures for Batch Jobs" on page 604.
Note: If you delete an object from the Hub Console (for example, if you
delete a mapping), the Batch Group tool highlights any batch jobs that depend
on that object (for example, a stage job) in red. You must resolve this issue
prior to re-executing the batch group.
Execution Paths
An execution path is the sequence in which batch jobs are executed when the
entire batch group is executed. The execution path begins with the Start node
and ends with the End node. The Batch Group tool does not validate the
execution sequence for you—it is up to you to ensure that the execution
sequence is correct. For example, the Batch Group tool would not notify you of
an error if you incorrectly specified the Load job for a base object ahead of its
Stage job.
Levels
In a batch group, the execution path consists of a series of one or more levels
that are executed in sequence (see "Running Batch Jobs in Sequence" on page
497).
- 513 -
All batch jobs in the level must complete before the batch group proceeds to
the next task in the sequence.
Note: Because all of the batch jobs in a level are executed in parallel, none of
the batch jobs in the same level should have any dependencies. For example,
the Stage and Load jobs for a base object should be in separate levels that are
executed in the proper sequence. For more information, see "Running Batch
Jobs in Sequence" on page 497.
In addition to using the Batch Group tool, you can execute batch groups in the
following ways:
• Services Integration Framework (SIF) requests—Applications can
invoke the SIF ExecuteBatchGroupRequest request to execute batch groups
directly. For more information, see the Informatica MDM Hub Services
Integration Framework Guide.
• Stored procedures—Execute batch groups through stored procedures
using any job scheduling software (such as Tivoli, CA Unicenter, and so
on). For more information, see "Executing Batch Groups Using Stored
Procedures" on page 598.
- 514 -
1. Start the Batch Group tool. For more information, see "Starting the Batch
Group Tool" on page 514.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Right-click the Batch Groups node in the Batch Group tree and choose Add
Batch Group from the pop-up menu.
The Batch Group tool adds a “New Batch Group” to the Batch Group tree.
Note the empty execution sequence. You will configure this after adding
the new batch group. For more information, see "Configuring Levels for
Batch Groups" on page 516.
4. Specify the following information:
Field Description
Name Specify a unique, descriptive name for this batch group.
Description Enter a description for this batch group.
- 515 -
3. In the navigation tree, expand the Batch Group node to show the batch
group that you want to delete.
4. Right-click the batch group that you want to delete, and then click Delete
Batch Group.
The Batch Group tool prompts you to confirm deletion.
5. Click Yes.
The Batch Group tool removes the deleted batch group from the navigation
tree.
As described in "About Batch Groups" on page 512, a batch group contains one
or more levels that are executed in sequence. This section describes how to
specify the execution sequence by configuring the levels in a batch group.
Command Description
Add Level Above Add a level to this batch group above the selected
item.
Add Level Below Add a level to this batch group below the selected item.
Move Level Up Move this batch group level above the prior level.
Move Level Down Move this batch group level below the next level.
Remove this Level Remove this batch group level.
The Batch Group tool displays the Choose Jobs to Add to Batch Group
dialog.
- 516 -
5. Expand the base object(s) for the job(s) that you want to add.
- 517 -
8. Click the button to save your changes.
5. Click Yes.
The Batch Group tool removes the deleted level from the batch group.
- 518 -
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. In the navigation tree, expand the Batch Group node to show the batch
group that you want to configure.
4. In the batch groups tree, right click on the level you want to move up, and
choose Move Level Up.
The Batch Group tool moves the level up within the batch group.
In the Batch Group tool, a job is a Informatica MDM Hub batch job. Each level
contains one or more batch jobs. If a level contains multiple batch jobs, then
all of those batch jobs are executed in parallel.
- 519 -
5. Expand the base object(s) for the job(s) that you want to add.
When configuring a batch group, you can configure job options for certain
kinds of batch jobs. For more information about these job options, see
- 520 -
"Options to Set Before Executing Batch Jobs" on page 504.
- 521 -
1. Start the Batch Group tool. For more information, see "Starting the Batch
Group Tool" on page 514.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. In the navigation tree, expand the Batch Group node to show the batch
group that you want to configure.
4. In the batch group, right click on the job that you want to move up, and
choose Move job down.
The Batch Group tool moves the selected job down one level in the batch
group.
Important: You must have the application server running for the duration of
an executing batch group.
Note: If you delete an object from the Hub Console (for example, if you
delete a mapping), the Batch Group tool highlights any batch jobs that depend
on that object (for example, a stage job) in red. You must resolve this issue
prior to re-executing the batch group.
The Control & Logs screen is where you can control the execution of a batch
group and view its execution logs.
- 522 -
3. Expand the batch group and click the Control & Logs node.
The Batch Group tool displays the Control & Logs screen for this batch
group.
Component Description
Toolbar Command buttons for managing batch group execution. For more
information, see "Command Buttons for Batch Groups" on page
523.
Logs for the Execution logs for this batch group.
Batch
Group
Logs for Execution logs for individual batch jobs in this batch group.
Batch Jobs
- 523 -
Button Description
Removes the selected group or job execution log.
For more information, see "Navigating to the Control & Logs Screen" on
page 522.
2. Click on the node and then select Batch Group > Execute, or click on the
Execute button.
The Batch Group tool executes the batch group and updates the logs panel
with the status of the batch group execution.
3. Click the Refresh button to see the execution result.
The Batch Group tool displays progress information.
- 524 -
Note: When you execute a batch group in FAILED status, you are actually re-
executing the failed instance, and the status is set to whatever the final
outcome is, and the Hub does not generate a new group log. However, in the
detailed logs (lower log table), you are not re-executing the failed instance,
rather, you are executing the same job in a new instance, and as a result, the
Hub generates a new log that is displayed here.
Each time that it executes a batch group, the Batch Group tool generates a
group execution log entry. Each log entry has the following properties:
Field Description
Status Current status of this batch job. If batch group execution failed,
displays a description of the problem. For more information, see
"Group Execution Status" on page 525.
Start Date / time when this batch job started.
End Date / time when this batch job ended.
Message Any messages regarding batch group execution.
- 525 -
Viewing the Job Execution Log for a Batch Job
Each time that it executes a batch job within a batch group, the Batch Group
tool generates a job execution log entry.
Note: If you want to view the metrics for a completed batch job, you can use
the Batch Viewer. For more information, see "Viewing Job Execution Logs" on
page 506.
If batch group execution fails, then you can resolve any problems that may
have caused the failure to occur, then restart batch group from the beginning.
- 526 -
3. Resolve any problems that may have caused the failure to occur and
execute the batch group again. For more information, see "Executing a
Batch Group" on page 524.
The Batch Group tool executes the batch group and creates a new
execution log entry.
Note: If a batch group fails and you do not click either the Set to Restart
button (see "Restarting a Batch Group That Failed Execution" on page 526) or
the Set to Incomplete button (see "Handling Incomplete Batch Group
Execution" on page 527) in the Logs for My Batch Group list, Informatica MDM
Hub restarts the batch job from the prior failed level.
In very rare circumstances, you might want to change the status of a running
batch group.
• If the batch group status says it is still executing, you can click Set Status
to Incomplete and execute the batch group again. You do this only if the
batch group has stopped executing (due to an error, such as a server
reboot or crash) but Informatica MDM Hub has not detected that the batch
group has stopped due to a job application lock in the metadata.
You will know this is a problem if the current status is Executing but the
database, application server, and logs show no activity. If this occurs,
click this button to clear the job application lock so that you can run the
batch group again; otherwise, you will not be able to execute the batch
group. Setting the status to Incomplete just updates the status of the batch
group (as well as all batch jobs within the batch group)—it does not
terminate processing.
Note that, if the job status is Incomplete, you cannot set the job status to
Restart.
• If the job status is Failed, you can click Set to Restart. Note that, if the
job status is Restart, you cannot set the job status to Incomplete.
Changing the status allows you to continue doing something else while the
batch group completes.
- 527 -
To set the status of a running batch group to incomplete:
1. In the Logs for My Batch Group list, select the execution log entry for the
running batch group that you want to mark as incomplete.
3. Execute the batch group again. For more information, see "Executing a
Batch Group" on page 524.
Note: If a batch group fails and you do not click either the Set to Restart
button (see "Restarting a Batch Group That Failed Execution" on page 526) or
the Set to Incomplete button (see "Handling Incomplete Batch Group
Execution" on page 527) in the Logs for My Batch Group list, Informatica MDM
Hub restarts the batch job from the prior failed level.
If batch group execution resulted in records being written to the rejects table
(during the execution of Stage jobs or Load jobs), then the job execution log
enables the View Rejects button.
- 528 -
2. Navigate and inspect the rejected records as needed.
3. Click Close.
3. Click the particular batch group log entry you want to review in the upper
half of the logs panel.
Informatica MDM Hub displays the detailed job execution logs for that
batch group in the lower half of the panel. For additional information, see:
• "Group Execution Status" on page 525
• "Viewing the Group Execution Log for a Batch Group" on page 525
• "Viewing the Job Execution Log for a Batch Job" on page 526
Note: Batch group logs can be deleted by selecting a batch group log and
clicking the Clear Selected button. To delete all logs shown in the panel,
click the Clear All button.
- 529 -
Deleting Batch Groups
To delete a batch group:
1. Start the Batch Group tool. For more information, see "Starting the Batch
Group Tool" on page 514.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. In the navigation tree, expand the Batch Group node to show the batch
group that you want to delete.
4. In the batch group, right click on the job that you want to move up, and
choose Delete Batch Group (or select Batch Group > Delete Batch
Group).
- 530 -
Batch Job Description
Tokens Jobs" strings that encode the columns used to identify candidates for
on page 540 matching.
"Hub Delete Deletes data from the Hub based on base object / XREF level
Jobs" on input.
page 541
"Key Match Matches records from two or more sources when these sources
Jobs" on use the same primary key. Compares new records to each other
page 541 and to existing records, and identifies potential matches based
on the comparison of source record keys as defined by the
match rules.
"Load Jobs" Copies records from a staging table to the corresponding target
on page 542 base object in the Hub Store. During the load process, applies
the current trust and validation rules to the records.
"Manual Link Shows logs for records that have been manually linked in the
Jobs" on Merge Manager tool. Used with link-style base objects only.
page 545
"Manual Shows logs for records that have been manually merged in the
Merge Jobs" Merge Manager tool. Used with merge-style base objects only.
on page 545
"Manual Shows logs for records that have been manually unlinked in the
Unlink Jobs" Merge Manager tool. Used with link-style base objects only.
on page 546
"Manual Shows logs for records that have been manually unmerged in
Unmerge the Merge Manager tool.
Jobs" on
page 546
"Match Jobs" Finds duplicate records in the base object, based on the current
on page 547 match rules.
"Match Conducts a search to gather match statistics but does not
Analyze actually perform the match process. If areas of data with the
Jobs" on potential for huge match requirements are discovered,
page 550 Informatica MDM Hub moves the records to a hold status, which
allows a data steward to review the data manually before
proceeding with the match process.
"Match for For data with a high percentage of duplicate records, compares
Duplicate new records to each other and to existing records, and identifies
Data Jobs" exact duplicates. The maximum number of exact duplicates is
on page 552 based on the Duplicate Match Threshold setting for this base
object.
"Migrate Link Used with link-style base objects only. Migrates link-style base
Style To objects to merge-style base objects.
Merge Style
Jobs" on
page 552
"Multi Merge Allows the merge of multiple records in one job.
Jobs" on
page 552
"Promote Reads the PROMOTE_IND column from an XREF table and
Jobs" on changes to ACTIVE the state on all rows where the column’s
page 552 value is 1.
"Recalculate Recalculates all base objects identified by ROWID_OBJECT
BO Jobs" on column in the table/inline view if you include the ROWID_
page 554 OBJECT_TABLE parameter.
If you do not include the parameter, this batch job recalculates
all records in the base object, in batches of MATCH_BATCH_
SIZE or 1/4 the number of the records in the table, whichever is
less.
- 531 -
Batch Job Description
"Recalculate Recalculates the BVT for the specified ROWID_OBJECT.
BVT Jobs" on
page 555
"Reset Links Updates the records in the _LINK table to account for changes in
Jobs" on the data. Used with link-style base objects only.
page 555
"Reset Match Shows logs of the operation where all matched records have
Table Jobs" been reset to be queued for match.
on page 555
"Revalidate Executes the validation logic/rules for records that have been
Jobs" on modified since the initial validation during the Load Process.
page 556
"Stage Jobs" Copies records from a landing table into a staging table. During
on page 556 execution, cleanses the data according to the current cleanse
settings.
"Synchronize Updates metadata for base objects. Used after a base object
Jobs" on has been loaded but not yet merged, and subsequent trust
page 557 configuration changes (such as enabling trust) have been made
to columns in that base object. This job must be run before
merging data for this base object.
Autolink Jobs
For link-style base objects only, after the Match job has been run, you can run
the Autolink job to automatically link any records that qualified for autolinking
during the match process.
- 532 -
until the maximum number of records for manual consolidation limit is
reached (see "Maximum Matches for Manual Consolidation" on page 368). The
match batch size parameter (see "Number of Rows per Match Job Batch Cycle"
on page 368) controls the number of records per cycle that this process goes
through to finish the match and merge cycles. For more information, see
"Match Jobs" on page 547 and "Automerge Jobs" on page 534.
Important: Do not run an Auto Match and Merge job on a base object that is
used to define relationships between records in inter-table or intra-table
match paths. Doing so will change the relationship data, resulting in the loss of
the associations between records. For more information, see "Relationship
Base Objects" on page 374.
If you execute an Auto Match and Merge job, it completes successfully with
one job shown in the status. However, if you stop and restart the application
server and return to the Batch Viewer, you see a second job (listed under
Match jobs) with a warning a few seconds later. The second job is to ensure
that either the base object is empty or there are no more records to match.
After running an Auto Match and Merge job, the Batch Viewer displays the
following metrics (if applicable) in the job execution log.
Metric Description
Matched Number of records that were matched by the Auto Match and
records Merge job.
Records Number of records that were tokenized prior to the Auto Match
tokenized and Merge job.
Automerged Number of records that were merged by the Auto Match and
records Merge job.
Accepted as Number of records that were accepted as unique records by the
unique Auto Match and Merge job. For more information, see
records "Automerge Jobs" on page 534.
Applies only if this base object has Accept All Unmatched
Rows as Unique enabled (set to Yes) in the Match / Merge
Setup configuration. For more information, see "Accept All
Unmatched Rows as Unique" on page 369.
- 533 -
Metric Description
Queued for Number of records that were queued for automerge by a Match
automerge job that was executed by the Auto Match and Merge job. For
more information, see "Automerge Jobs" on page 534.
Queued for Number of records that were queued for manual merge. Use the
manual Merge Manager in the Hub Console to process these records. For
merge more information, see the Informatica MDM Hub Data Steward
Guide.
Automerge Jobs
For merge-style base objects only, after the Match job has been run, you can
run the Automerge job to automatically merge any records that qualified for
automerging during the match process. When an Automerge job is run, it
processes all matches in the MATCH table that are flagged for automerging
(Automerge_ind=1).
Note: For state-enabled objects only, records that are PENDING (source and
target records) or DELETED are never automerged. When a record is deleted,
it is removed from the match table and its consolidation_ind is reset to 4. For
more information regarding how to manage the state of base object or XREF
records, refer to "Configuring State Management for Base Objects" on page
162.
Auto Match and Merge batch jobs execute a continual cycle of a Match job,
followed by an Automerge job, until there are no more records to match, or
until the maximum number of records for manual consolidation limit is
reached (see "Maximum Matches for Manual Consolidation" on page 368). For
additional information, see "Auto Match and Merge Jobs" on page 532.
Automerge Metrics
After running an Automerge job, the Batch Viewer displays the following
metrics (if applicable) in the job execution log:
- 534 -
Metric Description
Automerged Number of records that were automerged by the Automerge job.
records
Accepted as Number of records that were accepted as unique records by the
unique Automerge job. Applies only if this base object has Accept All
records Unmatched Rows as Unique enabled (set to Yes) in the Match
/ Merge Setup configuration. For more information, see "Accept
All Unmatched Rows as Unique" on page 369.
Note: For state-enabled base objects only, the BVT logic uses the HUB_
STATE_IND to ignore the non contributing base objects where the HUB_
STATE_IND is -1 or 0 (PENDING or DELETED state). For the online BUILD_BVT
call, provide INCLUDE_PENDING_IND parameter.
For more information regarding how to manage the state of base object or
XREF records, refer to "State Management" on page 159
- 535 -
External Match jobs can process both fuzzy-match and exact-match rules, and
can be used with fuzzy-match and exact-match base objects. For more
information, see "Exact-match and Fuzzy-match Base Objects" on page 247.
In addition to the base object and its associated match key table, the External
Match job uses the following input and output tables.
Each base object has an External Match Input (EMI) table for External Match
jobs. This table uses the following naming pattern:
C_BaseObject_EMI
where BaseObject is the name of the base object associated with this External
Match job.
When you create a base object, the Schema Manager automatically creates
the associated EMI table, and automatically adds the following system
columns:
- 536 -
Column Data Size Not Description
Name Type Null
SOURCE_ VARCHAR 50 Used as part of a three-column composite
KEY primary key to uniquely identify this record and to
map to records in the C_BaseObject_EMO table.
SOURCE_ VARCHAR 50 Used as part of a three-column composite
NAME primary key to uniquely identify this record and to
map to records in the C_BaseObject_EMO table.
FILE_ VARCHAR 50 Used as part of a three-column composite
NAME primary key to uniquely identify this record and to
map to records in the C_BaseObject_EMO table.
When populating the EMI table (see "Populating the Input Table" on page 539),
at least one of these columns must contain data. Note that the column names
are non-restrictive—they can contain any identifying data, as long as the
composite three-column primary key is unique.
In addition, when you configure match rules for a particular column (for
example, Person_Name, Address_Part1, or Exact_Cust_ID), the Schema
Manager adds that column automatically to the C_BaseObject_EMI table.
You can view the columns of an external match table in the Schema Manager
by expanding the External Match Table node.
The records in the EMI table are analogous to the match batch used in Match
jobs. As described in "Flagging the Match Batch" on page 251, the match batch
contains the set of records that are matched against the rest of records in the
base object. The difference is that, for Match jobs, the match batch records
reside in the base object, while for External Match, these records reside in a
separate input table.
Each base object has an External Match Output (EMO) table that contains the
output data for External Match jobs. This table uses the following naming
pattern:
C_BaseObject_EMO
where BaseObject is the name of the base object associated with this External
Match job.
- 537 -
Before the External Match job is executed, Informatica MDM Hub drops and
re-creates this table.
Instead of populating the match table for the base object, the External Match
job populates this EMO table with match pairs. Each row in the EMO represents
a pair of matched records—one from the EMI table and one from the base
object:
• The primary key (SOURCE_KEY + SOURCE_NAME + FILE_NAME) uniquely
identifies the record in the EMI table.
• ROWID_OBJECT_MATCHED uniquely identifies the record in the base
object.
- 538 -
Populating the Input Table
Before running an External Match job, the EMI table must be populated with
records to match against the records in the base object. The process of
loading data into an EMI table is external to Informatica MDM Hub—you must
use a data loading tool that works with your database platform (such as
SQL*Loader).
Important: When you populate this table, you must supply data for at least
one of the system columns (SOURCE_KEY, SOURCE_NAME, and FILE_NAME) to
help link back from the _EMI table. In addition, the C_BaseObject_EMI table
must contain flat records—like the output of a JOIN, with unique source keys
and no foreign keys to other tables.
- 539 -
Note: The C_BaseObject_EMO table is dropped and recreated after every
External Match Job execution.
The match process depends on the match tokens in the match key table being
current. If match tokens need to be updated (for example, if records have
been added or updated during the load process), the match process
automatically runs the tokenize process at the start of a match job (see
"Regenerating Match Tokens If Needed" on page 251). To expedite the match
process, it is recommended that you run the tokenize process separately—
before the match process—either by:
• manually executing the Generate Match Tokens job, or
• configuring the tokenize process to run automatically after the completion
of the load process (see "Generating Match Tokens (Optional)" on page
239)
For state-enabled base objects only, the tokenize process skips records that
are in the DELETED state. These records can be tokenized through the
Tokenize API, but will be ignored in batch processing. PENDING records can be
matched on a per-base object basis by setting the MATCH_PENDING_IND
(default off). For more information about how to manage the state of base
object or XREF records, see "Configuring State Management for Base Objects"
on page 162.
Before you run a Generate Match Tokens job, you can use the Re-generate
All Match Tokens check box to specify the scope of match token generation.
- 540 -
Do one of the following:
• Check (select) this check box to have the Generate Match Tokens job
tokenize all records in the base object.
• Uncheck (clear) this check box to have the Generate Match Tokens job
generate match tokens for only new or updated records in the base object
(whose DIRTY_IND=1, as described in "Base Object Records Flagged for
Tokenization" on page 243).
After the match tokens are generated, you can run the Match job for the base
object.
Note: Hub Delete jobs execute as a batch only stored procedure—you can not
call a Hub Delete job from the Batch Viewer or Batch Group tools, and there is
no corresponding SIF request that external applications can invoke. For more
information, see "Hub Delete Jobs" on page 575.
- 541 -
A Key Match job is automatically created after a primary key match rule for a
base object has been created or changed in the Schema Manager (Match /
Merge Setup configuration). For more information, see "Configuring Primary
Key Match Rules" on page 434.
Load Jobs
Load jobs move data from a staging table to the corresponding target base
object in the Hub Store. Load jobs also calculate trust values for base objects
with defined trusted columns, and they apply validation rules (if defined) to
determine the final trust values. For more information about loading data,
including trust, validation, and delta detection, see "Configuration Tasks for
Loading Data" on page 343.
For state-enabled base objects, the load batch process can load records in any
state. The state is specified as an input column on the staging table. The input
state can be specified in the mapping view a landing table column or it can be
derived. If an input state is not specified in the mapping, then the state is
assumed to be ACTIVE. For more information regarding how to manage the
state of base object or XREF records, refer to "Configuring State Management
for Base Objects" on page 162.
The following table describes how input states affect the states of existing
XREFs.
Existing ACTIVE PENDING DELETED No XREF No
XREF (Load by Base
State: rowid) Object
Incoming
XREF
State:
ACTIVE Update Update + Update + Insert Insert
Promote Restore
PENDING Pending Pending Pending Pending Pending
Update Update Update + Update Insert
Restore
DELETED Soft Hard Hard Delete Error Error
Delete Delete
Undefined Treat Treat as Treat as Treat As Treat
as Pending Deleted Active As
Active Active
The following table provides a matrix of how Informatica MDM Hub processes
records (for state-enabled base objects) during Load (and Put) for certain
operations based on the record state:
- 542 -
Incoming Existing Notes
Record Record
State State
Update the ACTIVE ACTIVE
XREF record
when:
DELETED ACTIVE
PENDING PENDING
ACTIVE PENDING
DELETED DELETED
PENDING DELETED
DELETED When a base object rowid delete record
comes in, Informatica MDM Hub updates the
base object and all XREF records (regardless
of ROWID_SYSTEM) to DELETED state.
Insert the PENDING ACTIVE The second record for the pair is created.
XREF record
when:
ACTIVENo
Record
PENDING No
Record
Delete the ACTIVE PENDING Delete the ACTIVE record in the pair, the
XREF record (for PENDING record is then updated.
when: paired Paired records are two records with the same
records) PKEY_SRC_OBJECT and ROWID_SYSTEM.
DELETED PENDING
Informatica PENDING ACTIVE Paired records are two records with the same
MDM Hub (for PKEY_SRC_OBJECT and ROWID_SYSTEM.
displays an paired
error when: records)
Additional notes:
• If the incoming state is not specified (for a Load update), then the
incoming state is assumed to be the same as the current state. For
example if the incoming state is null and the existing state of the XREF or
base object to update is PENDING, then the incoming state is assumed to
be PENDING instead of null.
• Informatica MDM Hub deletes XREF records using the Hub Delete batch
job. The Hub Delete batch job removes specified data—up to and including
an entire source system—from Informatica MDM Hub based on your base
object/XREF input to the cmxdm.hub_delete_batch stored procedure. For
more information, see "Hub Delete Jobs" on page 575.
For more information regarding how to manage the state of base object or
XREF records, refer to "Configuring State Management for Base Objects" on
page 162.
- 543 -
• Run a Load job only if the Stage job that loads the staging table used by
the Load job has completed successfully.
• Run the Load job for a parent table before you run the Load job for a child
table.
• If a lookup on the child object is not defined (the lookup table and column
were not populated), in order to successfully load data, you must repeat
the Stage job on the child object prior to running the Load job.
• Only one Load job at a time can be run for the same base object. Multiple
Load jobs for the same base object cannot be run concurrently.
Before you run a Load job, you can use the Force Update check box to
configure how the Load job loads data from the staging table to the target
base object. By default, Informatica MDM Hub checks the Last Update Date for
each record in the staging table to ensure that it has not already loaded the
record. To override this behavior, check (select) the Force Update check
box, which ignores the Last Update Date, forces a refresh, and loads each
record regardless of whether it might have already been loaded from the
staging table. Use this approach prudently, however. Depending on the
volume of data to load, forcing updates can carry a price in processing time.
When configuring the advanced properties of a base object in the Schema tool,
you can check (select) the Generate Match Tokens on Load check box to
generate match tokens during Load jobs, after the records have been loaded
into the base object. By default, this check box is unchecked (cleared), and
match tokens are generated during the Match process instead. For more
information, see "Editing Base Object Properties" on page 95 and "Run-time
Execution Flow of the Load Process" on page 231.
After running a Load job, the Batch Viewer displays the following metrics (if
applicable) in the job execution log.
- 544 -
Metric Description
Total Number of records processed by the Load job.
records
Inserted Number of records inserted by the Load job into the target
object.
Updated Number of records updated by the Load job in the target object.
No action Number of records on which no action was taken (the records
already existed in the base object).
Updated Number of records that updated the cross-reference table for
XREF this base object. If you are loading a record during an
incremental load, that record has already been consolidated
(exists only in the XREF and not in the base object).
Records Number of records tokenized by the Load job. Applies only if the
tokenized Generate Match Tokens on Load check box is selected in the
Schema tool. For more information, see "Generating Match
Tokens During Load Jobs" on page 544.
Merge Number of updated cross-reference records that have been
contributor merged into other rowid_objects. Represents the difference
XREF between the total number of updated cross-reference records
records and the number of updated base object records.
Missing Number of source records that were missing lookup information
Lookup / or had invalid rowid_object records.
Invalid
rowid_
object
records
- 545 -
Maximum Matches for Manual Consolidation
In the Schema Manager, you can configure the maximum number of matches
ready for manual consolidation to prevent data stewards from being
overwhelmed with thousands of manual merges for processing. Once this limit
is reached, the Match jobs and the Auto Match and Merge jobs will not run until
the number of matches has been reduced. For more information, see
"Maximum Matches for Manual Consolidation" on page 368.
When you start a Manual Merge job, the Merge Manager displays a dialog with
a progress indicator. A manual merge can take some time to complete. If
problems occur during processing, an error message is displayed on
completion. This error also shows up in the job execution log for the Manual
Merge job in the Batch Viewer.
In the Merge Manager, the process dialog includes a button labeled Mark
process as incomplete that updates the status of the Manual Merge job but
does not abort the Manual Merge job. If you click this button, the merge
process continues in the background. At this point, there will be an entry in the
Batch Viewer for this process. When the process completes, the success or
failure is reported. For more information about the Merge Manager, see the
Informatica MDM Hub Data Steward Guide.
When you start a Manual Unmerge job, the Data Manager displays a dialog
with a progress indicator. A manual unmerge can take some time to complete,
- 546 -
especially when a record in question is the product of many constituent
records If problems occur during processing, an error message is displayed on
completion. This error also shows up in the job execution log for the Manual
Unmerge in the Batch Viewer.
In the Data Manager, the process dialog includes a button labeled Mark
process as incomplete that updates the status of the Manual Unmerge job
but does not abort the Manual Unmerge job. If you click this button, the
unmerge process continues in the background. At this point, there will be an
entry in the Batch Viewer for this process. When the process completes, the
success or failure is reported.
Match Jobs
A match job generates search keys for a base object, searches through the
data for match candidates (records that are possible matches), applies the
match rules to the match candidates, generates the matches, and then queues
the matches for either automatic or manual consolidation. For an introduction,
see "Match Process" on page 245.
When you create a new base object in an ORS, Informatica MDM Hub
automatically creates its Match job. Each Match job compares new or updated
records in a base object with all records in the base object. For a detailed
description, see "Run-Time Execution Flow of the Match Process" on page 251.
After running a Match job, the matched rows are flagged for automatic and
manual consolidation. Informatica MDM Hub creates jobs that automatically
consolidate the appropriate records (automerge or autolink). If a record is
flagged for manual consolidation (manual merge or manual link), data
stewards must use the Merge Manager to perform the manual consolidation.
For more information about manual consolidation, see the Informatica MDM
Hub Data Steward Guide. For more information about consolidation, see
"About the Consolidate Process" on page 255.
You configure Match jobs in the Match / Merge Setup node in the Schema
Manager. For more information, see "Configuration Tasks for the Match
Process" on page 363.
Important: Do not run a Match job on a base object that is used to define
relationships between records in inter-table or intra-table match paths. Doing
so will change the relationship data, resulting in the loss of the associations
between records. For more information, see "Relationship Base Objects" on
page 374.
- 547 -
Match Tables
When a Informatica MDM Hub Match job runs for a base object, it populates its
match table with pairs of matched records. Match tables are usually named
Base_Object_MTCH. For more information, see "Populating the Match Table
with Match Pairs" on page 252.
The following table describes the details of the match batch process behavior
given the incoming states for state-enabled base objects:
Source Target Operation Result
Base Base
Object Object
State State
ACTIVE ACTIVE The records are analyzed for matching
PENDING ACTIVE Whether PENDING records are ignored in Batch Match is a
table-level parameter. If set, then batch match will
include PENDING records for the specified Base Object.
But the PENDING records can only be the source record in
a match.
DELETED Any state DELETED records are ignored in Batch Match
ANY PENDING PENDING records cannot be the target of a match.
Note: For Build Match Group (BMG), do not build groups with PENDING
records. PENDING records to be left as individual matches. PENDING matches
will have automerge_ind=2. For more information regarding how to manage
the state of base object or XREF records, refer to "Configuring State
Management for Base Objects" on page 162.
For merge-style base objects only, you can run the Auto Match and Merge job
for a base object. Auto Match and Merge batch jobs execute a continual cycle
of a Match job, followed by an Automerge job, until there are no more records
to match, or until the maximum number of records for manual consolidation
limit is reached (see "Maximum Matches for Manual Consolidation" on page
368). For more information, see "Auto Match and Merge Jobs" on page 532.
- 548 -
Jobs" on page 540. For more information about tokens for match, see
"About the Consolidate Process" on page 255.
• the Generate Match Tokens job need not be scheduled. Informatica MDM
Hub automatically runs it.
The Match job for a base object does not attempt to match every record in the
base object against every other record in the base object. Instead, you specify
(in the Schema tool):
• how many records the job should match each time it runs. For more
information, see "Number of Rows per Match Job Batch Cycle" on page
368.
• how many matches are allowed for manual consolidation.
This feature helps to prevent data stewards from being overwhelmed with
manual merges for processing. Once this limit is reached, the Match job
will not run until the number of matches ready for manual consolidation
has been reduced. For more information, see "Maximum Matches for
Manual Consolidation" on page 368.
For Match jobs, before executing the job, you can select the match rule set
that you want to use for evaluating matches.
The default match rule set for this base object is automatically selected. To
choose any other match rule set, click the drop-down list and select any other
match rule set that has been defined for this base object. For more
information, see "Configuring Match Rule Sets" on page 399.
After running a Match job, the Batch Viewer displays the following metrics (if
applicable) in the job execution log:
- 549 -
Metric Description
Matched Number of records that were matched by the Match job.
records
Records Number of records that were tokenized by the Match job.
tokenized
Queued Number of records that were queued for automerge by the Match
for job. Use the Automerge job to process these records. For more
automerge information, see "Automerge Jobs" on page 534.
Queued Number of records that were queued for manual merge by the
for manual Match job. Use the Merge Manager in the Hub Console to process
merge these records. For more information, see the Informatica MDM
Hub Data Steward Guide.
Each Match Analyze job is dependent on new / updated records in the base
object that have been tokenized and are thus queued for matching. For base
objects that have intertable match enabled, the Match Analyze job is also
dependent on the successful completion of the data tokenization jobs for all
child tables, which in turn is dependent on successful Load jobs for the child
tables.
You can limit the number of records that the Match Analyze job moves to the
on-hold status. By default, no limit is set. To configure a limit, edit the
cmxcleanse.properties file and add the following setting:
cmx.server.match.threshold_to_move_range_to_hold = n
- 550 -
where n is the maximum number of records that the Match Analyze job can
move to the on-hold status. For more information about the
cmxcleanse.properties file, see the Informatica MDM Hub Installation Guide.
After running a Match Analyze job, the Batch Viewer displays the following
metrics (if applicable) in the job execution log.
Metric Description
Records Number of records that were tokenized by the Match Analyze
tokenized job.
Records Number of records that were moved to a “Hold” status
moved to (consolidation indicator = 9) to avert overmatching. These
hold status records typically represent a hot spot in the data and are not run
through the match process. Data stewards can remove the hold
status in the Data Manager.
Records Number of records that were analyze for matching.
analyzed (to
be matched)
Match Number of actual matches that would be required to process this
comparisons base object.
required
Statistics
Statistic Description
Top 10 range count Top ten number of records in a given search range.
Top 10 range Top ten number of match comparison that will need to be
comparison count performed for a given search range.
Total records Count of the records moved to hold.
moved to hold
Total matches Total number of matches these records moved to hold
moved to hold required.
Total ranges Number of ranges required to process all the matches in
processed base object.
Total candidates Total number of match candidates required to process all
matches for this base object.
Time for analyze Amount of time required to run the analysis.
- 551 -
Match for Duplicate Data Jobs
Match for Duplicate Data jobs search for exact duplicates to consider them
matched. The maximum number of exact duplicates is based on the base
object columns defined in the Duplicate Match Threshold property in the
Schema Manager for each base object. For more information, see "Duplicate
Match Threshold" on page 91. For more information, see also "Matching for
Duplicate Data" on page 249.
Note: The Match for Duplicate Data job does not display in the Batch Viewer
when the duplicate match threshold is set to 1 and non-equal matches are
enabled on the base object.
Promote Jobs
For state-enabled objects, the Promote job reads the PROMOTE_IND column
from an XREF table and changes the system state to ACTIVE for all rows
where the column’s value is 1. Informatica MDM Hub resets PROMOTE_IND
after the Promote job has run.
- 552 -
Here are the behavior details for the Promote batch job:
XREF Base Object Hub Hub Refresh Resulting Operation Result
State State Before Action Action BVT? BO State
Before Promote on XREF on BO
Promote
PENDING ACTIVE Promote Update Yes ACTIVE Informatica MDM Hub
promotes the pending
XREF and recalculates the
BVT to include the
promoted XREF.
PENDING PENDING Promote Promote Yes ACTIVE Informatica MDM Hub
promotes the pending
XREF and base object.
The BVT is then calculated
based on the promoted
XREF.
DELETED This operation None None No The state Informatica MDM Hub
behaves the of the ignores DELETED records
same way resulting in Batch Promote. This
regardless of base scenario can only happen
the state of object if a record that had been
the base record is flagged for promotion is
object record. unchanged deleted prior to running
by this the Promote batch
operation. process.
ACTIVE This operation None None No The state Informatica MDM Hub
behaves the of the ignores ACTIVE records in
same way resulting Batch Promote. This
regardless of base scenario can only happen
the state of object if a record that had been
the base record is flagged for promotion is
object record. unchanged made ACTIVE prior to
by this running the Promote
operation. batch process.
You can run the Promote job using the following methods:
• Using the Hub Console; for more information, see "Running Promote Jobs
Using the Hub Console" on page 553.
• Using the CMXSM.AUTO_PROMOTE stored procedure; for more
information, see "Promote Jobs" on page 592.
• Using the Services Integration Framework (SIF) API (and the associated
SiperianClient Javadoc); for more information, see the Informatica MDM
Hub Services Integration Framework Guide.
- 553 -
• Batch Viewer according to the instructions in "Starting the Batch
Viewer Tool" on page 501
• Batch Group according to the instructions in "Starting the Batch Group
Tool" on page 514
2. Select the Promote job for the desired base object.
3. Execute the Promote job according to the instructions in "Running Batch
Jobs Manually" on page 502 or "Executing Batch Groups Using the Batch
Group Tool" on page 522.
4. Display the results of the Promote job according to the instructions in
"Viewing Job Execution Logs" on page 506.
After running a Promote job, the Batch Viewer displays the following metrics
(if applicable) in the job execution log.
Metric Description
Autopromoted records Number of records that were promoted by the
Promote job.
Deleted XREF records Number of XREF records that were deleted by the
Promote job.
Active records not Number of ACTIVE records that were not
promoted promoted.
Protected records not Number of protected records that were not
promoted promoted.
Once the Promote job has run, you can view these statistics on the job
summary page in the Batch Viewer.
Recalculate BO Jobs
There are two versions of Recalculate BO:
- 554 -
• Using the ROWID_OBJECT_TABLE Parameter—Recalculates all base
objects identified by ROWID_OBJECT column in the table/inline view (note
that brackets are required around inline view).
• Without the ROWID_OBJECT_TABLE Parameter—Recalculates all
records in the base object, in batches of MATCH_BATCH_SIZE or 1/4 the
number of the records in the table, whichever is less.
If you change your match rules after matching, you are prompted to reset
your matches. When you reset matches, everything in the match table is
deleted. In addition, the Reset Match Table job then resets the consolidation_
ind=4 where it is =2. For more information, see "About the Consolidate
Process" on page 255.
When you save changes to the schema match columns, the following message
box is displayed.
Click Yes to reset the existing matches and create a Reset Match Table job in
the Batch Viewer.
Note: If you do not reset the existing matches, your next Match job will take
longer to execute because Informatica MDM Hub will need to regenerate the
match tokens before running the Match job.
- 555 -
Revalidate Jobs
Revalidate jobs execute the validation logic/rules for records that have been
modified since the initial validation during the Load Process. You can run
Revalidate if/when records change post the initial Load process’s validation
step. If no records change, no records are updated. If some records have
changed and get caught by the existing validation rules, the metrics will show
the results.
Revalidate is executed manually using the batch viewer for base objects. For
more information, see "Running Batch Jobs Using the Batch Viewer Tool" on
page 501.
Stage Jobs
Stage jobs move data from a landing table to a staging table, performing any
cleansing that has been configured in the Informatica MDM Hub mapping
between the tables (see "Mapping Columns Between Landing and Staging
Tables" on page 286). Stage jobs have parallel cleanse jobs that you can run
(see "About Data Cleansing in Informatica MDM Hub" on page 307). The stage
status indicates which Cleanse Match Server is hit during a stage. For more
information about staging data, see "Configuration Tasks for the Stage
Process" on page 274.
Note: If the Stage job is grayed out, then the mapping has become invalid
due to changes in the staging table, in a column mapping, or in a cleanse
function. Open the specific mapping using the Mappings tool, verify it, and
then save it. For more information, see "Mapping Columns Between Landing
and Staging Tables" on page 286.
- 556 -
• You can run multiple Stage jobs simultaneously if there are multiple
Cleanse Match Servers set up to run the jobs.
After running a Stage job, the Batch Viewer displays the following metrics in
the job execution log.
Metric Description
Total Number of records processed by the Stage job.
records
Inserted Number of records inserted by the Stage job into the target object.
Rejected Number of records rejected by the Stage job. For more information,
see "Viewing Rejected Records" on page 510.
Synchronize Jobs
You must run the Synchronize job after any changes are made to the schema
trust settings. The Synchronize job is created when any changes are made to
the schema trust settings, as described in "Batch Jobs That Are Created When
Changes Occur" on page 500. For more information, see "Configuring Trust for
Source Systems" on page 344.
When you save changes to schema column trust settings in the Systems and
Trust tool, a message box reminds you to run the synchronize process.
- 557 -
Running Synchronize Jobs
To run the Synchronize job, navigate to the Batch Viewer, find the correct
Synchronize job for the base object, and run it. Informatica MDM Hub updates
the metadata for the base objects that have trust enabled after initial load has
occurred.
- 558 -
Chapter 18: Writing Custom Scripts
to Execute Batch Jobs
This chapter explains how to create custom scripts to execute batch jobs and
batch groups in a Informatica MDM Hub implementation. The information in
this chapter is intended for implementation teams and system administrators.
For information how to configure and execute Informatica MDM Hub batch
jobs using the Batch Viewer and Batch Group tools in the Hub Console, see
"About Informatica MDM Hub Batch Jobs" on page 496.
Important: You must have the application server running for the duration of
a batch job.
Chapter Contents
• "About Executing Informatica MDM Hub Batch Jobs" on page 559
• "Setting Up Job Execution Scripts" on page 560
• "Monitoring Job Results and Statistics" on page 563
• "Stored Procedure Reference" on page 566
• "Executing Batch Groups Using Stored Procedures" on page 598
• "Developing Custom Stored Procedures for Batch Jobs" on page 604
In the Hub Console, the Informatica MDM Hub Batch Viewer and Batch Group
tools provide simple mechanisms for executing Informatica MDM Hub batch
jobs. However, they do not provide a means for executing and managing jobs
on a scheduled basis. To execute and manage jobs according to a schedule,
you need to execute stored procedures that do the work of batch jobs or batch
groups. Most organizations have job management tools that are used to
control IT processes. Any such tool capable of executing Oracle PL*SQL or DB2
SQL commands can be used to schedule and manage Informatica MDM Hub
batch jobs.
- 559 -
Setting Up Job Execution Scripts
This section describes how to set up job execution scripts for running
Informatica MDM Hub stored procedures.
- 560 -
C_REPOS_TABLE_OBJECT_V has the following columns:
C_REPOS_TABLE_OBJECT_V Columns
Column Description
Name
ROWID_ Uniquely identifies a batch job.
TABLE_
OBJECT
ROWID_ Depending on the type of batch job, this is the table identifier for
TABLE either the table affected by the job (target table) or the table
providing the data for the job (source table).
• For Stage jobs, ROWID_TABLE refers to the target table
(staging table).
• For Load jobs, ROWID_TABLE refers to the source table
(staging table).
• For Match, Match Analyze, Autolink, Automerge, Auto Match
and Merge, External Match, Generate Match Tokens, and Key
Match jobs, ROWID_TABLE refers to the base object table,
which is both source and target for the jobs.
OBJECT_ Description of the type of batch job. Examples include:
NAME • Stage jobs: CMX_CLEANSE.EXE.
• Load jobs: CMXLD.LOAD_MASTER.
• Match and Match Analyze jobs: CMXMA.MATCH.
OBJECT_ Description of the batch job, including the type of batch job as
DESC well as the object affected by the batch job. Examples include:
• Stage for C_STG_CUSTOMER_CREDIT
• Load from C_STG_CUSTOMER_CREDIT
• Match and Merge for C_CUSTOMER
OBJECT_ Together with OBJECT_FUNCTION_TYPE_CODE, this is a foreign
TYPE_CODE key to C_REPOS_OBJ_FUNCTION_TYPE.
An OBJECT_TYPE_CODE of “P” indicates a procedure that can
potentially be executed by a scheduling tool.
OBJECT_ Indicates the actual procedure type (stage, load, match, and so
FUNCTION_ on).
TYPE_CODE
PUBLIC_ Indicates whether the procedure is a procedure that can be
IND displayed in the Batch Viewer.
PARAMETER Describes the parameter list for the procedure. Where specific
ROWID_TABLE values are required for the procedure, these are
shown in the parameter list. Otherwise, the name of the
parameter is simply displayed in the parameter list.
An exception to this is the parameter list for Stage jobs (where
OBJECT_NAME = CMX_CLEANSE.EXE). In this case, the full
parameter list is not shown. For a list of parameters, see "Stage
Jobs" on page 596.
VALID_IND If VALID_IND is not equal to 1, do not execute the
procedure. It means that some repository settings have
changed that affect the procedure. This usually applies to
changes that affect the Stage jobs if the mappings have not been
checked and saved again. For more information, see
"Determining Available Execution Scripts" on page
563."Determining Available Execution Scripts" on page 563
- 561 -
Identifiers in the C_REPOS_TABLE_OBJECT_V View
- 562 -
OBJECT_NAME OBJECT_DESC OBJECT_ OBJECT_ OBJECT_
TYPE_ FUNCTION_ FUNCTION_
CODE TYPE_CODE TYPE_DESC
CMXUT.SYNC Synchronize after changes are made to P S Synchronize
the schema trust settings.
CMXMM.UNMERGE Unmerge for BaseObjectName P X Manual
unmerge
- 563 -
Returned Description
Parameter
OUT_ERROR_ Error message if an error occurred.
MSG
OUT_RETURN_ Return code. Zero (0) if no errors occurred, or one (1) if an
CODE error occurred.
Error handling code in job execution scripts can look for return codes and trap
any associated error messaged.
• Certain stored procedures will also register the temporary tables that
remain so that a server-side process can periodically remove them.
- 564 -
The following table describes the various repository tables.
- 565 -
Table Description
Name
Note: All the input parameters that need a delimited list require a trailing “~”
character.
- 566 -
Batch Job Description
Jobs" on
page 575
"Hub Delete Deletes data from the Hub based on base object / XREF level
Jobs" on input.
page 575
"Key Match Matches records from two or more sources when these sources
Jobs" on use the same primary key. Compares new records to each other
page 579 and to existing records, and identifies potential matches based
on the comparison of source record keys as defined by the
match rules.
"Load Jobs" Copies records from a staging table to the corresponding target
on page 580 base object in the Hub Store. During the load process, it also
applies the current trust and validation rules to the records.
"Manual Link Shows logs for records that have been manually linked in the
Jobs" on Merge Manager tool. Used with link-style base objects only.
page 581
"Manual Shows logs for records that have been manually merged in the
Merge Jobs" Merge Manager tool. Used with merge-style base objects only.
on page 581
"Manual Shows logs for records that have been manually unlinked in the
Unlink Jobs" Data Manager tool. Used with link-style base objects only.
on page 582
"Manual Shows logs for records that have been manually unmerged in
Unmerge the Merge Manager tool. Used with merge-style base objects
Jobs" on only.
page 583
"Match Jobs" Finds duplicate records in the base object, based on the current
on page 586 match rules.
"Match Conducts a search to gather match statistics but does not
Analyze actually perform the match process. If areas of data with the
Jobs" on potential for huge match requirements are discovered,
page 588 Informatica MDM Hub moves the records to a hold status, which
allows a data steward to review the data manually before
proceeding with the match process.
"Match for For data with a high percentage of duplicate records, compares
Duplicate new records to each other and to existing records, and identifies
Data Jobs" exact duplicates. The maximum number of exact duplicates is
on page 589 based on the Duplicate Match Threshold setting for this base
object.
Note: The Match for Duplicate Data batch job has been
deprecated.
"Promote Reads the PROMOTE_IND column from an XREF table and
Jobs" on changes to ACTIVE the state on all rows where the column’s
page 592 value is 1.
"Recalculate Recalculates all base objects identified by ROWID_OBJECT
BO Jobs" on column in the table/inline view if you include the ROWID_
page 592 OBJECT_TABLE parameter.
If you do not include the parameter, this batch job recalculates
all records in the base object, in batches of MATCH_BATCH_
SIZE or 1/4 the number of the records in the table, whichever is
less.
"Recalculate Recalculates the BVT for the specified ROWID_OBJECT.
BVT Jobs" on
page 593
- 567 -
Batch Job Description
"Reset Batch Resets a batch group. For more information, see "Stored
Group Status Procedures for Batch Groups" on page 599.
Jobs" on
page 594
"Reset Links Updates the records in the _LINK table to account for changes in
Jobs" on the data. Used with link-style base objects only.
page 594
"Reset Match Shows logs of the operation where all matched records have
Table Jobs" been reset to be queued for match.
on page 594
"Revalidate Executes the validation logic/rules for records that have been
Jobs" on modified since the initial validation during the Load Process. You
page 595 can run Revalidate if/when records change after the initial Load
process’s validation step. If no records change, no records are
updated. If some records have changed and get caught by the
existing validation rules, the metrics will show the results.
"Stage Jobs" Copies records from a landing table into a staging table. During
on page 596 execution, cleanses the data according to the current cleanse
settings.
"Synchronize Updates metadata for base objects. Used after a base object
Jobs" on has been loaded but not yet merged, and subsequent trust
page 597 configuration changes (such as enabling trust) have been made
to columns in that base object. This job must be run before
merging data for this base object.
- 568 -
)
SELECT ROWID_USER
INTO V_ROWID_USER
FROM C_REPOS_USER
WHERE USER_NAME = 'ADMIN';
- 569 -
Autolink Jobs
Autolink jobs automatically link records that have qualified for autolinking
during the match process and are flagged for autolinking (Autolink_ind = 1).
Important: Do not run an Auto Match and Merge job on a base object that is
used to define relationships between records in inter-table or intra-table
match paths. Doing so will change the relationship data, resulting in the loss of
the associations between records. For more information, see "Relationship
Base Objects" on page 374.
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
The Auto Match and Merge jobs for a target base object can either be run on
successful completion of each Load job, or on successful completion of all
Load jobs for the object.
Auto Match and Merge jobs must complete with a RUN_STATUS of 0
(Completed Successfully) or 1 (Completed with Errors) to be considered
successful.
- 570 -
)
Sample Job Execution Script for Auto Match and Merge Jobs
DECLARE
IN_ROWID_TABLE CHAR(14);
IN_USER_NAME VARCHAR2(50);
IN_MATCH_SET_NAME VARCHAR(500);
OUT_ERROR_MSG VARCHAR2(1024);
OUT_RETURN_CODE NUMBER;
BEGIN
IN_ROWID_TABLE := 'SVR1.188';
IN_USER_NAME := 'CMX_ORS';
IN_MATCH_SET_NAME := 'MRS2';
OUT_ERROR_MSG := NULL;
OUT_RETURN_CODE := NULL;
CMXMA.MATCH_AND_MERGE ( IN_ROWID_TABLE, IN_USER_NAME,
IN_MATCH_SET_NAME, OUT_ERROR_MSG, OUT_RETURN_CODE );
DBMS_OUTPUT.Put_Line('OUT_ERROR_MESSAGE = ' || OUT_ERROR_MESSAGE);
DBMS_OUTPUT.Put_Line('RC = ' || TO_CHAR(OUT_RETURN_CODE));
COMMIT;
END;
Automerge Jobs
Automerge jobs automatically merge records that have qualified for
automerging during the match process and are flagged for automerging
(Automerge_ind = 1). Automerge jobs are used with merge-style base objects
only. For more information, see "Automerge Jobs" on page 534.
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
- 571 -
,OUT_ERROR_MESSAGE OUT VARCHAR2(1024) --Error message, if any
,OUT_RETURN_CODE OUT NUMBER --Return code (if no errors, 0 is returned)
)
BEGIN
IN_ROWID_TABLE := NULL;
IN_USER_NAME := NULL;
OUT_ERROR_MESSAGE := NULL;
OUT_RETURN_CODE := NULL;
For more information, see "Stored Procedures for Batch Groups" on page 599.
- 572 -
Note: The External Batch job executes as a batch job only—there is no
corresponding SIF request that external applications can invoke.
You should run Generate Match Tokens jobs whenever match tokens need to
be regenerated, as described in "When to Generate Match Tokens" on page
242. Generate Match Tokens jobs apply to fuzzy-match base objects only—not
to exact-match base objects—as described in "Exact-match and Fuzzy-match
- 573 -
Base Objects" on page 247. For more information, see "Generate Match
Tokens Jobs" on page 540.
Note: The Generate Match Tokens job generates the match tokens for the
entire base object (when IN_FULL_RESTRIP_IND is set to 1). Check (select)
the Re-generate All Match Tokens check box in the Batch Viewer to populate
the IN_FULL_RESTRIP_IND parameter.
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
- 574 -
IN_FULL_RESTRIP_IND := NULL;
For more information, see "Stored Procedures for Batch Groups" on page 599.
Although the Hub Delete job deletes the XREF record, a pointer to the deleted
record (actually to the parent base object of this XREF) could potentially be
present on the
_HMXR table (on column ORIG_TGT_ROWID_OBJECT). The Match Tree tool
displays REMOVED (ID#: xxxx) for the removed record(s).
Important:
• The Hub Delete batch job will not delete the data if there are records
queued for an Automerge job.
• Do not run a Hub Delete job when there are automerge records in the
match table. Run the Hub Delete job after the automerge matches are
processed.
Cascade Delete
The Hub Delete job performs a cascade delete if you set the parameter IN_
ALLOW_CASCADE_DELETE_IND=1 for a base object in the stored procedure.
With cascade delete, when records in the parent object are deleted, Hub
Delete also removes the affected records in the child base object. Hub Delete
checks each child base object table for related data that should be deleted
given the removal of the parent base object record.
- 575 -
Important: For the prior example, the Hub Delete job may potentially delete
XREF records from other source systems. To ensure that Hub Delete does not
delete XREF records from other systems, do not use cascade delete. IN_
ALLOW_CASCADE_DELETE_IND forces Hub Delete to delete the child base
objects and cross-references (regardless of system) when the parent base
object is being deleted.
Notes:
• If you do not set the IN_ALLOW_CASCADE_DELETE_IND=1, Informatica
MDM Hub generates an error message if there are child base objects
referencing the deleted base objects record; Hub Delete fails, and
Informatica MDM Hub performs a rollback operation for the associated
data.
• IN_CASCADE_CHILD_SYSTEM_XREF=1 is not supported in XU SP1. Since
there may be situations where you would want to selectively cascade
deletes to child records, you would have to perform child deletes first, and
then parent deletes with the cascade delete feature disabled.
Note: Informatica MDM Hub sets the HUB_STATE_IND to -9 in the HXRF when
XREFs are deleted. The HIST table will be set to -9 if the base object record is
deleted.
The Hub Delete job removes “records on hold” or records that have had their
CONSOLIDATION_IND column set to 9.
- 576 -
,OUT_RETURN_CODE OUT INT
,OUT_TMP_TABLE_LIST IN OUT VARCHAR2(32000)
,IN_RECALCULATE_BVT IN INT DEFAULT 1
,IN_ALLOW_CASCADE_DELETE IN INT DEFAULT 1
,IN_CASCADE_CHILD_SYSTEM_XREF IN INT DEFAULT 0
,IN_OVERRIDE_HISTORY_IND IN INT DEFAULT 0
,IN_PURGE_HISTORY_IND IN INT DEFAULT 0
,IN_USER_NAME IN VARCHAR2(50) DEFAULT NULL
,IN_ALLOW_COMMIT_IND IN INT DEFAULT 1
)
Parameters
Parameter Description
IN_BO_TABLE_ Name of the table that contains the list of base objects to delete.
NAME
IN_XREF_ Name of the table that contains the list of XREFs to delete.
LIST_TO_BE_
DELETED
IN_ If set to one (1), recalculates BVT following base object and/or
RECALCULATE_ XREF delete.
BVT_IND
IN_ALLOW_ If set to one (1), specifies that when records in the parent object
CASCADE_ are deleted, Hub Delete also removes the affected records in the
DELETE_IND child base object. Hub Delete checks each child base object table
for related data that should be deleted given the removal of the
parent base object record.
IN_CASCADE_ Not supported in XU SP1. Leave the value for this parameter as
CHILD_ the default (0) when executing the procedure.
SYSTEM_XREF
IN_OVERRIDE_ If set to one (1), Hub Delete does not write to history tables
HISTORY_IND when deleting. If you set IN_OVERRIDE_HISTORY_IND=1 and set IN_
PURGE_HISTORY_IND=1, then Hub Delete removes history tables to
delete all traces of the data.
IN_PURGE_ If set to one (1), Hub Delete will remove all history records
HISTORY_IND related to deleted XREF records.
Returns
Parameter Description
OUT_ Number of deleted XREFs.
DELETED_
XREF_
COUNT
OUT_ Number of deleted base objects.
DELETED_
BO_
COUNT
OUT_TMP_ List of delimited tables that can be passed on to CMXUT.DROP_
TABLE_ TEMP_TABLES stored procedure calls to clean up the temporary
LIST tables. For more information, see "Housekeeping for Temporary
Tables" on page 564.
OUT_ERROR_ Error message text.
MSG
OUT_ Error code. If zero (0), then the stored procedure completed
RETURN_ successfully.
- 577 -
Parameter Description
CODE The procedure will return a non-zero value in case of an error.
DELETE TMP_DELETE_KEYS;
- 578 -
COMMIT;
END; /
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
Key Match jobs are dependent on the successful completion of the Load job
responsible for loading data into the base object. The Key Match job cannot
have been run after any changes were made to the data.
- 579 -
CMXMA.KEY_MATCH (IN_ROWID_TABLE, IN_USER_NAME, OUT_ERROR_MESSAGE, OUT_
RETURN_CODE);
DBMS_OUTPUT.Put_Line(' Row id table = ' || IN_ROWID_TABLE);
CMXMA.KEY_MATCH ( IN_ROWID_TABLE, IN_USER_NAME, OUT_ERROR_MESSAGE, OUT_
RETURN_CODE);
DBMS_OUTPUT.Put_Line('OUT_ERROR_MESSAGE = ' || OUT_ERROR_MESSAGE);
DBMS_OUTPUT.Put_Line('OUT_RETURN_CODE = ' || TO_CHAR(OUT_RETURN_CODE));
COMMIT;
END;
Load Jobs
Load jobs move data from staging tables to the final target objects, and apply
any trust and validation rules where appropriate. For more information about
Load jobs and the load process, see "Load Jobs" on page 542.
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
Each Load job is dependent on the success of the Stage job that precedes it. In
addition, each Load job is governed by the demands of referential integrity
constraints and is dependent on the successful completion of all other Load
jobs responsible for populating tables referenced by the base object that is the
target of the load. Run the loads for parent tables before the loads for child
tables.
- 580 -
Sample Job Execution Script for Load Jobs
DECLARE
IN_STG_ROWID_TABLE CHAR(14);
IN_USER_NAME VARCHAR2(50);
OUT_ERROR_MSG VARCHAR2(1024);
OUT_RETURN_CODE NUMBER;
IN_FORCE_UPDATE_IND NUMBER;
IN_ROWID_JOB_GRP_CTRL CHAR(14);
IN_ROWID_JOB_GRP_ITEM CHAR(14);
BEGIN
IN_STG_ROWID_TABLE := 'SVR1.1L9';
IN_USER_NAME := 'ADMIN';
IN_ROWID_JOB_GRP_CTRL := NULL;
IN_ROWID_JOB_GRP_ITEM := NULL;
OUT_ERROR_MSG := NULL;
OUT_RETURN_CODE := NULL;
IN_FORCE_UPDATE_IND := 1;
CMXLD.LOAD_MASTER ( IN_STG_ROWID_TABLE, IN_USER_NAME, OUT_ERROR_MSG,
OUT_RETURN_CODE,IN_FORCE_UPDATE_IND, IN_ROWID_JOB_GRP_CTRL, IN_ROWID_JOB_
GRP_ITEM);
DBMS_OUTPUT.Put_Line('OUT_ERROR_MSG = ' || OUT_ERROR_MSG);
DBMS_OUTPUT.Put_Line('OUT_RETURN_CODE = ' || TO_CHAR(OUT_RETURN_CODE));
COMMIT;
END;
- 581 -
,IN_USER_NAME IN VARCHAR2(50)
,OUT_MERGED_IS_UNIQUE_IND OUT INT
,OUT_ERROR_MESSAGE OUT VARCHAR2(1024)
,OUT_RETURN_CODE OUT INT
,CALLED_MANUALLY_IND IN INT DEFAULT 1
,OUT_TMP_TABLE_LIST OUT NOCOPY VARCHAR2(32000)
)
- 582 -
Manual Unmerge Jobs
The Unmerge job can unmerge already-consolidated records, whether those
records were consolidated using Automerge, Manual Merge, manual edit, Load
by Rowid_Object, or Put Xref. The Unmerge job succeeds or fails as a single
transaction: if the server fails while the Unmerge job is executing, the
unmerge process is rolled back.
Cascade Unmerge
The Unmerge job performs a cascade unmerge if this feature is enabled for
this base object in the Schema Manager in the Hub Console. With cascade
unmerge, when records in the parent object are unmerged, Informatica MDM
Hub also unmerges affected records in the child base object.
For example: Customer A record (parent) in the Customer base object has
multiple address records (children) in the Address base object. The two tables
are linked by a unique key (Customer_ID).
• When cascade unmerge is enabled—Unmerging the parent record
(Customer A) in the Customer base object also unmerges Customer A's
child address records in the Address base object.
• When cascade unmerge is disabled—Unmerging the parent record
(Customer A) in the Customer base object has no effect on Customer A's
child records in the Address base object; they are NOT unmerged.
In your job execution script, you can specify the scope of records to unmerge
by setting IN_UNMERGE_ALL_XREFS_IND.
• IN_UNMERGE_ALL_XREFS_IND=0: Default setting. Unmerges the single
record identified in the specified XREF to its state prior to the merge.
• IN_UNMERGE_ALL_XREFS_IND=1: Unmerges all XREFs to their state prior
to the merge. Use this option to quickly unmerge all XREFs for a single
consolidated record in a single operation.
- 583 -
Linear and Tree Unmerge
In your job execution script, you can specify the type of unmerge (linear or
tree unmerge) by setting IN_TREE_UNMERGE_IND:
• IN_TREE_UNMERGE_IND=0: Default setting. Linear Unmerge
• IN_TREE_UNMERGE_IND=1: Tree Unmerge
Linear Unmerge
Tree Unmerge
During a tree unmerge, you unmerge a tree of merged base object records as
an intact sub-structure. A sub-tree having unmerged base object records as
root will come out from the original merge tree structure. (For example,
merge a1 and a2 into a, then merge b1 and b2 into b, and then finally merge a
and b into c. If you then perform a tree unmerge on a, and then unmerge a
from a1, a2 is a sub tree and will come out from the original tree c. As a
result, a is the root of the tree after the unmerge.)
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
- 584 -
Committing Unmerge Transactions After Error Checking
Important: After calling the Unmerge, check the return code (OUT_RETURN_
CODE) in your error handling.
• If any failure occurred during execution (OUT_RETURN_CODE <> 0),
immediately roll back any changes. Wait until after you have successfully
rolled back the unmerge changes before you invoke Unmerge again.
• If no failure occurred during execution (OUT_RETURN_CODE = 0), commit
any unmerge changes.
Each Manual Unmerge job is dependent on data having already been merged.
Note: If the manual unmerge job completes successfully, you need to pass
the OUT_TMP_TABLE_LIST parameter to the CMXUT.DROP_TEMP_TABLES
stored procedure to clean up the temporary tables that were returned in the
parameter, as described in "Housekeeping for Temporary Tables" on page
564.
- 585 -
IN_PKEY_SRC_OBJECT VARCHAR2 (255);
IN_TREE_UNMERGE_IND NUMBER;
IN_ROWID_JOB_CTL CHAR (14);
IN_INTERACTION_ID NUMBER;
IN_USER_NAME VARCHAR2 (50);
OUT_UNMERGED_ROWID CHAR (14);
OUT_TMP_TABLE_LIST VARCHAR2 (32000);
OUT_ERROR_MESSAGE VARCHAR2 (1024);
RC NUMBER;
IN_UNMERGE_ALL_XREFS_IND NUMBER;
BEGIN
IN_ROWID_TABLE := 'SVR1.8ZC ';
IN_ROWID_SYSTEM := 'SVR1.7NJ ';
IN_PKEY_SRC_OBJECT := '6';
IN_TREE_UNMERGE_IND := 0; -- Default 0, 1 for tree unmerge
IN_ROWID_JOB_CTL := NULL;
IN_INTERACTION_ID := NULL;
IN_USER_NAME := 'XHE';
OUT_UNMERGED_ROWID := NULL;
OUT_TMP_TABLE_LIST := NULL;
OUT_ERROR_MESSAGE := NULL;
RC := NULL;
IN_UNMERGE_ALL_XREFS_IND := 0; -- default 0, 1 for unmerge_all
Match Jobs
Match jobs find duplicate records in the base object, based on the current
match rules. For more information about Match jobs and the match process,
see "Match Jobs" on page 547.
Important: Do not run a Match job on a base object that is used to define
relationships between records in inter-table or intra-table match paths. Doing
so will change the relationship data, resulting in the loss of the associations
between records. For more information, see "Relationship Base Objects" on
page 374.
For a complete list of the identifiers used to execute the stored procedure
associated with this batch job, see "Identifiers in the C_REPOS_TABLE_
OBJECT_V View" on page 562.
- 586 -
Dependencies for Match Jobs
Each Match job is dependent on new / updated records in the base object that
have been tokenized and are thus queued for matching. For parent base
objects that have children, the Match job is also dependent on the successful
completion of the data tokenization jobs for all child tables, which in turn is
dependent on successful Load jobs for the child tables.
- 587 -
COMMIT;
END;
For a complete list of the identifiers used to execute the stored procedure
associated with this batch job, see "Identifiers in the C_REPOS_TABLE_
OBJECT_V View" on page 562.
Each Match Analyze job is dependent on new / updated records in the base
object that have been tokenized and are thus queued for matching. For parent
base objects, the Match Analyze job is also dependent on the successful
completion of the data tokenization jobs for all child tables, which in turn is
dependent on successful Load jobs for the child tables.
- 588 -
BEGIN
IN_ROWID_TABLE := NULL;
IN_USER_NAME := NULL;
OUT_ERROR_MSG := NULL;
OUT_RETURN_CODE := NULL;
IN_VALIDATE_TABLE_NAME := NULL;
IN_MATCH_ANALYZE_IND := 1;
Note: The Match for Duplicate Data batch job has been deprecated.
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
Match for Duplicate Data jobs require the existence of unconsolidated data in
the base object.
- 589 -
,OUT_RETURN_CODE OUT INT --Return code (if no errors, 0 is returned)
)
Sample Job Execution Script for Match for Duplicate Data Jobs
DECLARE
IN_ROWID_TABLE CHAR(14);
IN_USER_NAME VARCHAR2(200);
OUT_ERROR_MSG VARCHAR2(2000);
OUT_RETURN_CODE NUMBER;
BEGIN
IN_ROWID_TABLE := NULL;
IN_USER_NAME := NULL;
OUT_ERROR_MSG := NULL;
OUT_RETURN_CODE := NULL;
- 590 -
Identifiers for Executing Multi Merge Jobs
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
Each Multi Merge job is dependent on the successful completion of the match
process for this base object.
BEGIN
IN_ROWID_TABLE := 'SVR1.CP4 ';
IN_SURVIVING_ROWID := '40 ';
IN_MEMBER_ROWID_LIST := '42 ~44 ~45 ~47
~48 ~49 ~';
- 591 -
IN_ROWID_MATCH_RULE := NULL;
IN_COL_LIST := 'SVR1.CSB ~SVR1.CSE ~SVR1.CSG ~SVR1.CSH
~SVR1.CSA ~';
IN_VAL_LIST := 'INDU~THOMAS~11111111111~F~1000~';
IN_INTERACTION_ID := 0;
IN_USER_NAME := 'INDU';
IN_WINNING_CELL_OVERRIDE := NULL
OUT_ERROR_MESSAGE := NULL;
OUT_RETURN_CODE := NULL;
Promote Jobs
For state-enabled objects, a promote job reads the PROMOTE_IND column
from an XREF table and for all rows where the column’s value is 1, changes
the ACTIVE state to on. Informatica MDM Hub resets PROMOTE_IND after the
Promote job has run. For more information regarding how to manage the state
of base object or XREF records, refer to "About State Management in
Informatica MDM Hub" on page 159.
Recalculate BO Jobs
There are two versions of Recalculate BO:
• Using the ROWID_OBJECT_TABLE Parameter—Recalculates all BOs
identified by ROWID_OBJECT column in the table/inline view (note that
brackets are required around inline view).
- 592 -
• Without the ROWID_OBJECT_TABLE Parameter—Recalculates all
records in the BO, in batches of MATCH_BATCH_SIZE or 1/4 the number of
the records in the table, whichever is less.
- 593 -
)
For more information, see "Stored Procedures for Batch Groups" on page 599.
Note: This job cannot be run from the Batch Viewer. For more information,
see "Reset Match Table Jobs" on page 555.
- 594 -
CMXMA.RESET_MATCH( V_ROWID_TABLE, 'ADMIN', OUT_ERROR_MESSAGE, OUT_RETURN_
CODE );
DBMS_OUTPUT.PUT_LINE( 'RETURN MESSAGE: ' || SUBSTR(
OUT_ERROR_MESSAGE, 1, 255 ));
DBMS_OUTPUT.PUT_LINE( 'RETURN CODE: ' || OUT_RETURN_CODE );
DBMS_OUTPUT.Put_Line('OUT_ERROR_MESSAGE = ' || OUT_ERROR_MESSAGE);
DBMS_OUTPUT.Put_Line('RC = ' || TO_CHAR(RC));
COMMIT;
END;
Revalidate Jobs
Revalidate jobs execute the validation logic/rules for records that have been
modified since the initial validation during the Load Process. You can run
Revalidate if/when records change post the initial Load process’s validation
step. If no records change, no records are updated. If some records have
changed and get caught by the existing validation rules, the metrics will show
the results. Revalidate is executed manually using the batch viewer for base
objects. For more information. see "Running Batch Jobs Using the Batch
Viewer Tool" on page 501.
Note: Revalidate can only be run after an initial load and prior to merge on
base objects that have validate rules setup.
CMXUT.REVALIDATE_BO(IN_TABLE_NAME, IN_TABLE_NAME,
OUT_ERROR_MESSAGE, RC);
DBMS_OUTPUT.PUT_LINE ( 'OUT_ERROR_MESSAGE= ' ||
SUBSTR(OUT_ERROR_MESSAGE,1,200) );
DBMS_OUTPUT.Put_Line('OUT_ERROR_MESSAGE = ' || OUT_ERROR_MESSAGE);
DBMS_OUTPUT.Put_Line('RC = ' || TO_CHAR(RC));
COMMIT;
END;
- 595 -
Stage Jobs
Stage jobs copy records from a landing to a staging table. During execution,
Stage jobs optionally cleanse data according to the current cleanse settings.
For more information about Stage jobs and the stage process, see "Stage
Jobs" on page 556.
To learn about the identifiers used to execute the stored procedure associated
with this batch job, see "Identifiers in the C_REPOS_TABLE_OBJECT_V View"
on page 562.
- 596 -
IN_STG_ROWID_TABLE := NULL;
IN_ROWID_TABLE_OBJECT := NULL;
IN_RUN_SYNCH := NULL;
OUT_ERROR_MSG := NULL;
OUT_ERROR_CODE := NULL;
SELECT A.ROWID_TABLE, A.ROWID_TABLE_OBJECT INTO IN_STG_ROWID_TABLE,
IN_ROWID_TABLE_OBJECT
FROM C_REPOS_TABLE_OBJECT_V A, C_REPOS_TABLE B
WHERE A.OBJECT_NAME = 'CMX_CLEANSE.EXE'
AND B.ROWID_TABLE = A.ROWID_TABLE
AND B.TABLE_NAME = 'C_HMO_ADDRESS'
AND A.VALID_IND = 1;
CMXCL.START_CLEANSE ( IN_STG_ROWID_TABLE, IN_USER_NAME, IN_ROWID_TABLE_
OBJECT, IN_RUN_SYNCH, OUT_ERROR_MSG, OUT_ERROR_CODE );
DBMS_OUTPUT.PUT_LINE(' MESSAGE IS = ' || OUT_ERROR_MSG);
DBMS_OUTPUT.Put_Line('OUT_ERROR_MESSAGE = ' || OUT_ERROR_MSG);
DBMS_OUTPUT.Put_Line('OUT_ERROR_CODE = ' || TO_CHAR(OUT_ERROR_CODE));
COMMIT;
END;
Synchronize Jobs
You must run the Synchronize job after any changes are made to the schema
trust settings. The Synchronize job is created when any changes are made to
the schema trust settings, as described in "Batch Jobs That Are Created When
Changes Occur" on page 500. For more information, see "Configuring Trust for
Source Systems" on page 344.
To run the Synchronize job, navigate to the Batch Viewer, find the correct
Synchronize job for the base object, and run it. Informatica MDM Hub updates
the metadata for the base objects that have trust enabled after initial load has
occurred. For more information, see "Synchronize Jobs" on page 557.
- 597 -
INTO V_ROWID_TABLE
FROM C_REPOS_TABLE
WHERE TABLE_NAME = 'C_CUSTOMER';
This section describes how to execute batch groups using stored procedures
and job scheduling software (such as Tivoli, CA Unicenter, and so on).
Informatica MDM Hub provides stored procedures for managing batch groups.
For more information, see "Stored Procedures for Batch Groups" on page 599.
You can also use the Batch Group tool in the Hub Console to configure and run
batch groups. However, to schedule batch groups, you need to do so using
stored procedures, as described in this section. For more information about
the Batch Group tool, see "Running Batch Jobs Using the Batch Group Tool" on
page 512.
Note: If a batch group fails and you do not click either the Set to Restart
button (see "Restarting a Batch Group That Failed Execution" on page 526) or
the Set to Incomplete button (see "Handling Incomplete Batch Group
Execution" on page 527) in the Logs for My Batch Group list, Informatica MDM
Hub restarts the batch job from the prior failed level.
- 598 -
Stored Procedures for Batch Groups
Informatica MDM Hub provides the following stored procedures for managing
batch groups:
Stored Procedure Description
CMXBG.EXECUTE_ Performs an HTTP POST to the SIF
BATCHGROUP ExecuteBatchGroupRequest. For more information, see
"CMXBG.EXECUTE_BATCHGROUP" on page 599
CMXBG.RESET_ Performs an HTTP POST to the SIF
BATCHGROUP ResetBatchGroupRequest. For more information, see
"CMXBG.RESET_BATCHGROUP" on page 601.
CMXBG.GET_ Performs an HTTP POST to the SIF
BATCHGROUP_ GetBatchGroupStatusRequest. For more information, see
STATUS "CMXBG.GET_BATCHGROUP_STATUS" on page 602.
In addition to using parameters that are associated with the corresponding SIF
request, these stored procedures require the following parameters:
• URL of the Hub Server (for example, http://localhost:7001/cmx/request)
• username and password
• target ORS
CMXBG.EXECUTE_BATCHGROUP
Execute Batch Group jobs execute a batch group. Execute Batch Groups jobs
have an option to execute asynchronously, but not to receive a JMS response
for asynchronous execution. If you need to use asynchronous execution and
need to know when execution is finished, then poll with the cmxbg.get_
batchgroup_status stored procedure. Alternatively, if you need to receive a
JMS response for asynchronous execution, then execute the batch group
directly in an external application (instead of a job execution script) by
invoking the SIF ExecuteBatchGroup request, which is described in the
Informatica MDM Hub Services Integration Framework Guide.
Signature
FUNCTION CMXBG.EXECUTE_BATCHGROUP(
IN_MRM_SERVER_URL IN VARCHAR2(500)
, IN_USERNAME IN VARCHAR2(500)
, IN_PASSWORD IN VARCHAR2(500)
, IN_ORSID IN VARCHAR2(500)
, IN_BATCHGROUP_UID IN VARCHAR2(500)
, IN_RESUME IN VARCHAR2(500)
, IN_ASYNCRONOUS IN VARCHAR2(500)
, OUT_ROWID_BATCHGROUP_LOG OUT VARCHAR2(500)
, OUT_ERROR_MSG OUT VARCHAR2(500)
) RETURN NUMBER --Return the error code
- 599 -
Parameters
Name Description
IN_MRM_ Hub Server SIF URL.
SERVER_URL
IN_USERNAME User account with role-based permissions to execute batch
groups.
IN_PASSWORD Password for the user account with role-based permissions to
execute batch groups.
IN_ORSID ORS ID as shown in Console > Configuration >
Databases. For more information, see “Configuring Operational
Record Stores” on page 62.
IN_ Informatica MDM Hub Object UID of batch group to [execute,
BATCHGROUP_ reset, get status, etc.].
UID
IN_RESUME One of the following values:
• true: if previous execution failed, resume at that point
• false: regardless of previous execution, start from the
beginning
IN_ Specifies whether to execute asynchronously or
ASYNCRONOUS synchronously. One of the following values:
• true: start execution and return immediately
(asynchronous execution).
• false: return when group execution is complete
(synchronous execution).
Returns
Parameter Description
OUT_ROWID_ c_repos_job_group_control.rowid_job_group_control
BATCHGROUP_
LOG
OUT_ERROR_ Error message text.
MSG
NUMBER Error code. If zero (0), then the stored procedure completed
successfully. If one (1), then the stored procedure returns an
explanation in out_error_msg.
- 600 -
CMXLB.DEBUG_PRINT('EXECUTE_BATCHGROUP:
' || ' CODE='|| RET_VAL || ' MESSAGE='|| OUT_ERROR_MSG ||
' | OUT_ROWID_BATCHGROUP_LOG='|| OUT_ROWID_BATCHGROUP_LOG);
);
COMMIT;
END;
CMXBG.RESET_BATCHGROUP
Note: In addition to this stored procedure, there are Java API requests and
the SOAP and HTTP XML protocols available using Services Integration
Framework (SIF). The Reset Batch Group Status job has the following SIF API
requests available: ResetBatchGroup. For more information about this SIF API
request, see the Informatica MDM Hub Services Integration Framework
Guide.
Signature
FUNCTION CMXBG.RESET_BATCHGROUP(
IN_MRM_SERVER_URL IN VARCHAR2(500)
, IN_USERNAME IN VARCHAR2(500)
, IN_PASSWORD IN VARCHAR2(500)
, IN_ORSID IN VARCHAR2(500)
, IN_BATCHGROUP_UID IN VARCHAR2(500)
, OUT_ROWID_BATCHGROUP_LOG OUT VARCHAR2(500)
, OUT_ERROR_MSG OUT VARCHAR2(500)
) RETURN NUMBER --Return the error code
Parameters
Name Description
IN_MRM_ Hub Server SIF URL.
SERVER_URL
IN_USERNAME User account with role-based permissions to execute batch
groups.
IN_ Password for the user account with role-based permissions to
PASSWORD execute batch groups.
IN_ORSID ORS ID as specified in the Database tool in the Hub Console.
For more information, see "Configuring Operational Reference
Stores" on page 55.
IN_ Informatica MDM Hub Object UID of batch group to [execute,
BATCHGROUP_ reset, get status of, and so on].
UID
Returns
Parameter Description
OUT_ROWID_ c_repos_job_group_control.rowid_job_group_control
BATCHGROUP_
LOG
OUT_ERROR_ Error message text.
MSG
NUMBER Error code. If zero (0), then the stored procedure completed
- 601 -
Parameter Description
successfully. If one (1), then the stored procedure returns an
explanation in out_error_msg.
CMXBG.GET_BATCHGROUP_STATUS
Get Batch Group Status jobs return the batch group status.
Note: In addition to this stored procedure, there are Java API requests and
the SOAP and HTTP XML protocols available using Services Integration
Framework (SIF). The Get Batch Group Status job has the following SIF API
requests available: GetBatchGroupStatus. For more information about this SIF
API request, see the Informatica MDM Hub Services Integration Framework
Guide.
Signature
FUNCTION CMXBG.GET_BATCHGROUP_STATUS(
IN_MRM_SERVER_URL IN VARCHAR2(500)
, IN_USERNAME IN VARCHAR2(500)
, IN_PASSWORD IN VARCHAR2(500)
, IN_ORSID IN VARCHAR2(500)
, IN_BATCHGROUP_UID IN VARCHAR2(500)
, IN_ROWID_BATCHGROUP_LOG IN VARCHAR2(500)
, OUT_ROWID_BATCHGROUP OUT VARCHAR2(500)
, OUT_ROWID_BATCHGROUP_LOG OUT VARCHAR2(500)
, OUT_START_RUNDATE OUT VARCHAR2(500)
, OUT_END_RUNDATE OUT VARCHAR2(500)
, OUT_RUN_STATUS OUT VARCHAR2(500)
, OUT_STATUS_MESSAGE OUT VARCHAR2(500)
, OUT_ERROR_MSG OUT VARCHAR2(500)
) RETURN NUMBER --Return the error code
- 602 -
Parameters
Name Description
IN_MRM_ Hub Server SIF URL.
SERVER_URL
IN_USERNAME User account with role-based permissions to execute batch
groups.
IN_ Password for the user account with role-based permissions to
PASSWORD execute batch groups.
IN_ORSID ORS ID as specified in the Database tool in the Hub Console.
For more information, see "Configuring Operational Reference
Stores" on page 55.
IN_ Informatica MDM Hub Object UID of batch group to [execute,
BATCHGROUP_ reset, get status of, and so on].
UID If IN_ROWID_BATCHGROUP_LOG is null, the most recent log
for this group will be used.
IN_ROWID_ c_repos_job_group_control.rowid_job_group_control
BATCHGROUP_ Either IN_BATCHGROUP_UID or IN_ROWID_BATCHGROUP_
LOG LOG is required.
Returns
Parameter Description
OUT_ROWID_ c_repos_job_group.rowid_job_group
BATCHGROUP
OUT_ROWID_ c_repos_job_group_control.rowid_job_group_control
BATCHGROUP_
LOG
OUT_START_ Date / time when this batch job started.
RUNDATE
OUT_END_ Date / time when this batch job ended.
RUNDATE
OUT_RUN_ Job execution status code that is displayed in the Batch Group
STATUS tool. For more information, see "Executing Batch Groups
Using the Batch Group Tool" on page 522.
OUT_STATUS_ Job execution status message that is displayed in the Batch
MESSAGE Group tool. For more information, see "Executing Batch
Groups Using the Batch Group Tool" on page 522.
OUT_ERROR_ Error message text for this stored procedure call, if
MSG applicable.
NUMBER Error code. If zero (0), then the stored procedure completed
successfully. If one (1), then the stored procedure returns an
explanation in out_error_msg.
Sample Job Execution Script for Get Batch Group Status Jobs
DECLARE
OUT_ROWID_BATCHGROUP CMXLB.CMX_SMALL_STR;
OUT_ROWID_BATCHGROUP_LOG CMXLB.CMX_SMALL_STR;
OUT_START_RUNDATE CMXLB.CMX_SMALL_STR;
OUT_END_RUNDATE CMXLB.CMX_SMALL_STR;
OUT_RUN_STATUS CMXLB.CMX_SMALL_STR;
OUT_STATUS_MESSAGE CMXLB.CMX_SMALL_STR;
OUT_ERROR_MSG CMXLB.CMX_SMALL_STR;
OUT_RETURNCODE INT;
RET_VAL INT;
BEGIN
- 603 -
RET_VAL := CMXBG.GET_BATCHGROUP_STATUS(
'HTTP://LOCALHOST:7001/CMX/REQUEST/PROCESS/'
, 'ADMIN'
, 'ADMIN'
,'LOCALHOST-MRM-XU_3009'
, 'BATCH_GROUP.MYBATCHGROUP'
, NULL
, OUT_ROWID_BATCHGROUP
, OUT_ROWID_BATCHGROUP_LOG
, OUT_START_RUNDATE
, OUT_END_RUNDATE
, OUT_RUN_STATUS
, OUT_STATUS_MESSAGE
, OUT_ERROR_MSG
);
CMXLB.DEBUG_PRINT('GET_BATCHGROUP_STATUS: CODE='|| RET_VAL || '
MESSAGE='|| OUT_ERROR_MSG || ' STATUS=' || OUT_STATUS_MESSAGE || ' | OUT_
ROWID_BATCHGROUP_LOG='|| OUT_ROWID_BATCHGROUP_LOG);
END;
/
Signature
PROCEDURE EXAMPLE_JOB(
IN_ROWID_TABLE_OBJECT IN CHAR(14) --C_REPOS_TABLE_OBJECT.ROWID_TABLE_
OBJECT, RESULT OF CMXUT.REGISTER_CUSTOM_TABLE_OBJECT
,IN_USER_NAME IN VARCHAR2(50) --Username calling the function
- 604 -
,IN_ROWID_JOB IN CHAR(14) --C_REPOS_JOB_CONTROL.ROWID_JOB, for reference,
do not update status
,OUT_ERR_MSG OUT VARCHAR --Message about success or error
,OUT_ERR_CODE OUT INT -- >=0: Completed successfully. <0: Error
)
Parameters
Name Description
in_rowid_table_object IN c_repos_table_object.rowid_table_object
cmxlb.cmx_rowid Result of cmxut.REGISTER_CUSTOM_
TABLE_OBJECT
in_user_name IN User name calling the function.
cmxlb.cmx_user_name
Returns
Parameter Description
out_err_msg Error message text.
out_err_code Error code.
Signature
PROCEDURE REGISTER_CUSTOM_TABLE_OBJECT(
IN_ROWID_TABLE IN CHAR(14)
, IN_OBJ_FUNC_TYPE_CODE IN VARCHAR
, IN_OBJ_FUNC_TYPE_DESC IN VARCHAR
, IN_OBJECT_NAME IN VARCHAR
)
Parameters
Name Description
IN_ROWID_TABLE Foreign key to c_repos_table.rowid_table.
CMXLB.CMX_ROWID When the Hub Server calls the custom job in a batch
group, this value is passed in.
IN_OBJ_FUNC_TYPE_ Job type code. Must be 'A' for batch group custom jobs.
CODE
IN_OBJ_FUNC_TYPE_ Display name for the custom batch job in the Batch
DESC Groups tool in the Hub Console.
IN_OBJECT_NAME package.procedure name of the custom job.
- 605 -
Example
BEGIN
cmxut.REGISTER_CUSTOM_TABLE_OBJECT (
'SVR1.RS1B ' -- c_repos_table.rowid_table
,'A' -- Job type, must be 'A' for batch group
,'CMXBG_EXAMPLE.UPDATE_TABLE EXAMPLE' -- Display name
,'CMXBG_EXAMPLE.UPDATE_TABLE' -- Package.procedure
);
END;
Example
DECLARE
IN_ROWID_TABLE CHAR(14);
IN_ROWID_COL_LIST VARCHAR2(2000);
IN_USER_NAME VARCHAR2(50);
IN_INDEX_TYPE VARCHAR2(200);
BEGIN
IN_ROWID_TABLE := '<ROWID_TABLE>' ; -- rowid_table from c_repos_
table where table_name = 'your table name'
-- Notes:
-- 1. Trailing spaces in the rowid_column values are significant
-- 2. Separate each rowid_column with a ~ character and end the
list with ~ character e.g. '123 ~456 ~'
- 606 -
Removing Data from a Base Object and Supporting
Metadata Tables
Use the CMXUT.CLEAN_TABLE procedure to remove all data from a base object
and its supporting metadata tables. If a base object is referenced by a foreign
key in another base object, then the referencing base object must be empty
before you run cmxut.clean_table for the referenced base object.
Example
DECLARE
IN_TABLE_NAME VARCHAR2(30);
OUT_ERROR_MESSAGE VARCHAR2(1024);
RC NUMBER;
BEGIN
IN_TABLE_NAME := 'C_BO_TO_CLEAN'; --Name of the BO table
OUT_ERROR_MESSAGE := NULL; --Return msg; output parameter
RC := NULL; --Return code; output parameter
CMXUT.CLEAN_TABLE ( IN_TABLE_NAME, OUT_ERROR_MESSAGE, RC );
COMMIT;
END;
Example
DECLARE
IN_DEBUG_TEXT VARCHAR2(32000);
BEGIN
IN_DEBUG_TEXT := NULL; --String that you want to print in the log file
CMXLB.DEBUG_PRINT ( IN_DEBUG_TEXT );
COMMIT;
END;
PROCEDURE UPDATE_TABLE(
IN_ROWID_TABLE_OBJECT IN CMXLB.CMX_ROWID
,IN_USER_NAME IN CMXLB.CMX_USER_NAME
,IN_ROWID_JOB IN CMXLB.CMX_ROWID
,OUT_ERR_MSG OUT VARCHAR
,OUT_ERR_CODE OUT INT
- 607 -
);
END CMXBG_EXAMPLE;
/
CREATE OR REPLACE PACKAGE BODY CMXBG_EXAMPLE
AS
BEGIN
DECLARE
CUTOFF_DATE DATE;
RECORD_COUNT INT;
RUN_STATUS INT;
STATUS_MESSAGE VARCHAR2 (2000);
START_DATE DATE := SYSDATE;
MRM_ROWID_TABLE CMXLB.CMX_ROWID;
OBJ_FUNC_TYPE CHAR (1);
JOB_ID CHAR (14);
SQL_STMT VARCHAR2 (2000);
TABLE_NAME VARCHAR2(30);
RET_CODE INT;
REGISTER_JOB_ERR EXCEPTION;
BEGIN
SQL_STMT :=
'ALTER SESSION SET NLS_DATE_FORMAT=''DD MON YYYY
HH24:MI:SS''';
SELECT ROWID_TABLE
INTO MRM_ROWID_TABLE
FROM C_REPOS_TABLE_OBJECT
WHERE ROWID_TABLE_OBJECT = IN_ROWID_TABLE_OBJECT;
SELECT START_RUN_DATE
INTO CUTOFF_DATE
FROM C_REPOS_JOB_CONTROL
WHERE ROWID_JOB = IN_ROWID_JOB;
- 608 -
RECORD_COUNT := SQL%ROWCOUNT;
COMMIT;
-- For testing, sleep to make the procedure take longer
-- dbms_lock.sleep(5);
-- Set zero or many metrics about the job
CMXUT.SET_METRIC_VALUE (IN_ROWID_JOB, 1, RECORD_COUNT,
OUT_ERR_CODE, OUT_ERR_MSG);
COMMIT;
- 609 -
Part 5: Configuring Application Access
Contents
• "Generating ORS-specific APIs and Message Schemas" on page 611
• "Setting Up Security" on page 621
• "Viewing Registered Custom Code" on page 678
• "Auditing Informatica MDM Hub Services and Events" on page 684
- 610 -
Chapter 19: Generating ORS-specific
APIs and Message Schemas
This chapter describes how to use the SIF Manager tool to generate ORS-
specific APIs and how to use the JMS Event Schema Manager tool to generate
ORS-specific JMS Event Message objects.
Chapter Contents
• "Before You Begin" on page 611
• "Generating ORS-specific APIs" on page 611
• "Generating ORS-specific Message Schemas" on page 615
Note: Use of the ORS-specific API does not imply that you must use the SIF
SDK. Alternatively, you could use the ORS-specific API as SOAP web-services.
- 611 -
Starting the SIF Manager Tool
To start the SIF Manager tool:
1. In the Hub Console, connect to an Operational Reference Store (ORS). For
more information, see "Changing the Target Database" on page 37.
2. Expand the Informatica Utilities workbench and then click SIF Manager.
The Hub Console displays the SIF Manager tool.
- 612 -
Note: The following procedure assumes that you have already configured the
base objects and packages of the ORS. If you subsequently change any of
these, regenerate the ORS-specific APIs.
Note: SIF API generation requires that at least one secure package, remote
package, cleanse function, or mapping be defined.
Note: To prevent running out of heap space for the associated SIF API
Javadoc, you may need to increase the size of the heap. The default heap size
is 256M. You can override this default using the sif.jvm.heap.size parameter
in the cmxserver.properties file.
- 613 -
different from the logical name of any other ORS registered on this server
to prevent duplicates.
4. Click Generate and Deploy ORS-specific SIF APIs.
SIF Manager generates the APIs. The time this requires depends on the
size of the ORS schema. When the generation is complete, SIF Manager
deploys the ORS-specific web services and displays their URL. Note that
these are not required for Java ORS-specific APIs to work. Java and web
services ORS-specific APIs have no dependencies on each other, so you
can use one while the other is not in use.
You can use the resulting URL to access the WSDL descriptions from your
development environment.
Note: To prevent running out of heap space for the associated SIF API
Javadocs, you may need to increase the size of the heap. The default heap
size is 256M. You can also override this default using the sif.jvm.heap.size
parameter.
You can download ORS-specific JAR file at any point after the APIs have been
generated.
- 614 -
Finding Out-of-Sync Objects
The SIF Manager Find Out of Sync Objects function compares the last
generated APIs to the defined objects in the ORS. The SIF Manager reports
any differences between these. If differences are found, the ORS-specific API
should be regenerated.
Note: Once you have evaluated the impact of the out-of-sync objects, you can
then decide whether or not to re-generate the schema (typically, external
components which interact with the Hub are written to work with a specific
version of the generated schema). If you regenerate the schema, these
external components may no longer work.
Note: If your Informatica MDM Hub implementation requires that you use the
legacy XML message format (Informatica MDM Hub XU version) instead of the
current version of the XML message format (described in this section), see
"Legacy JMS Message XML Reference" on page 479 instead.
Use the JMS Event Schema Manager tool to generate and deploy ORS-specific
JMS Event Messages for the current ORS. The XML schema for these messages
- 615 -
can be downloaded or accessed using a URL. For more information about JMS
Event Messages, see "JMS Message XML Reference" on page 464.
Note: JMS Event Schema generation requires at least one secure package or
remote package be defined.
Important: If there are two databases that have the same schema (for
example, CMX_ORS), the logical name (which is the same as the schema
name) will be duplicated for JMS Events when the configuration is initially
saved. Consequently, the database display name is unique and should be used
as the initial logical name instead of the schema name to be consistent with
the SIF APIs. You will need to change the logical name before generating the
schema.
Additionally, each ORS has an XSD file specific to the ORS that uses the
elements from the common XSD file (siperian-mrm-events.xsd). The ORS-
specific XSD is named as <ors-name>-siperian-mrm-event.xsd. The XSD defines
two objects for each package and remote package in the schema:
Object Name Description
[packageName]Event Complex type containing elements of type
EventMetadata and [packageName].
[packageName]Record Complex type representing a package and its fields.
Also includes an element of type SipMetadata. This
complex type resembles the package record structures
defined in the Informatica MDM HubServices
Integration Framework (SIF). For more information,
refer to the Informatica MDM Hub Services Integration
Framework Guide.
- 616 -
The Hub Console displays the JMS Event Schema Manager tool.
The JMS Event Schema Manager tool displays the following areas:
Area Description
JMS ORS- Shows the event message schema for the ORS.
specific Use this function to generate and deploy ORS-specific JMS Event
Event Messages for the current ORS. The logical name is used to name
Message the components of the deployment. The schema can be
Schema downloaded or accessed using a URL.
Note: If legacy XML event message objects are to be used, ORS-
specific message object generation is not required.
Out of Sync Shows the database objects in the schema that are out of sync.
Objects with the generated API.
Important: If there are two databases that have the same schema (for
example, CMX_ORS), the logical name (which is the same as the schema
name) will be duplicated for JMS Events when the configuration is initially
saved. Consequently, the database display name is unique and should be used
as the initial logical name instead of the schema name to be consistent with
the SIF APIs. You will need to change the logical name before generating the
schema.
Additional notes:
- 617 -
• The following procedure assumes that you have already configured the
base objects, packages, and mappings of the ORS. If you subsequently
change any of these, regenerate the ORS-specific schemas.
• JMS Event Schema generation requires at least one secure package or
remote package.
An XSD file defines the structure of an XML file and can also be used to
validate the XML file. For example, if an XML file contains a reference to an
XSD, an XML validation tool can be used to verify that the tags in the XML
conform to the definitions defined in the XSD.
You use Find Out Of Sync Objects to determine if the event schema needs to
be re-generated to reflect changes in the system. The JMS Event Schema
Manager displays a list of packages and remote packages that have changed
since the last schema generation.
- 618 -
Note: The Out of Sync Objects function compares the generated APIs to the
database objects in the schema so both must be present to find the out-of-
sync objects.
Note: Once you have evaluated the impact of the out-of-sync objects, you can
then decide whether or not to re-generate the schema (typically, external
components which interact with the Hub are written to work with a specific
version of the generated schema). If you regenerate the schema, these
external components may no longer work.
If the JMS Event Schema Manager returns any out-of-sync objects, click
Generate and Deploy ORS-specific Schema to re-generate the event
schema. For more information, see "Generating and Deploying ORS-specific
Schemas" on page 617.
You can configure Informatica MDM Hub to periodically search for out-of-sync
objects and re-generate the schema as needed. This auto-poll feature
operates within the data change monitoring thread which automatically
engages a specified number of milliseconds between polls. You specify this
time frame using the Message Check Interval in the Message Queues tool.
When the monitoring thread is active, this automatic service first checks if the
out-of-sync interval has elapsed and if so, performs the out-of-sync check and
then re-generates the event schema as needed.
- 619 -
Note: If you bypass this step, the Hub issues a warning in the server log
asking you configure the schema generation.
2. Enable the Queue Status for Data Changes Monitoring message. For more
information, see "Configuring Global Message Queue Settings" on page
451.
3. Select the root node Message Queues and set the Out of sync check
interval (milliseconds). For more information, see "Configuring Global
Message Queue Settings" on page 451.
Since the out-of-sync auto-poll feature effectively depends on the Message
check interval, you should set the Out-of-sync check interval to a value
greater than or equal to that of the Message check interval.
Note: You can disable to out-of-sync check by setting the out-of-sync
check interval to 0.
- 620 -
Chapter 20: Setting Up Security
This chapter describes how to set up security for your Informatica MDM Hub
implementation using the Hub Console. To learn how to configure user access
to the Hub Console, see "About User Access to Hub Console Tools" on page
737.
Chapter Contents
• "About Setting Up Security" on page 621
• "Securing Informatica MDM Hub Resources" on page 629
• "Configuring Roles" on page 638
• "Configuring Informatica MDM Hub Users" on page 646
• "Configuring User Groups" on page 658
• "Assigning Users to the Current ORS Database" on page 661
• "Assigning Roles to Users and User Groups" on page 662
• "Managing Security Providers" on page 664
- 621 -
Security Access Manager
Authentication
- 622 -
Authorization
Informatica MDM Hub provides general types of resources that you can
configure to be secure resources: base objects, mappings, packages, remote
packages, cleanse functions, match rule sets, batch groups, metadata, content
metadata, Metadata Manager, HM profiles, the audit table, and the users
table. You can configure security for these resources in a highly granular way,
granting access to Informatica MDM Hub resources according to various
privileges (read, create, update, merge, and execute). Resources are either
PRIVATE (the default) or SECURE. Privileges can be granted only to secure
resources. To learn more see "Securing Informatica MDM Hub Resources" on
page 629.
Roles
- 623 -
Access to Hub Console Tools
For users who will be using the Hub Console to access Informatica MDM Hub
resources, you can use the Tool Access tool in the Configuration workbench to
control access privileges to Hub Console tools. For example, data stewards
typically have access to only the Data Manager and Merge Manager tools. For
more information, see "About User Access to Hub Console Tools" on page 737.
- 624 -
• a specific role is configured to have access to one or more secure
resources.
• each secure resource is configured with specific privileges (READ, WRITE,
CREATE, and so on) that define that role’s access to the secure resource.
• a user is assigned one or more roles.
At run time, in order to execute a SIF request, the logged-in user must be
assigned a role that has the required privilege(s) to access the resource(s)
involved with the request. Otherwise, the user’s request will be denied.
Internal-only PDP
The following figure shows a security deployment in which all PDPs are
handled internally by Informatica MDM Hub.
In this scenario, Informatica MDM Hub makes all policy decisions based on
how users, groups, roles, privileges, and resources are configured using the
Hub Console.
- 625 -
In this scenario, the external user directory manages user accounts, groups,
and user profiles. The external user directory is able to authenticate users and
provide information to Informatica MDM Hub about group membership and
user profile information.
Users or user groups that are maintained in the external user directory must
still be registered in Informatica MDM Hub. Registration is required so that
Informatica MDM Hub roles—and their associated privileges—can be assigned
to these users and groups.
In this scenario, external roles are explicitly mapped to Informatica MDM Hub
roles.
The following figure shows a security deployment in which role definition and
privilege assignment—in addition to user accounts, groups, user profiles, and
role assignment—is handled externally to Informatica MDM Hub.
- 626 -
In this scenario, Informatica MDM Hub simply exposes the protected
resources using external proxies, which are synchronized with the internally-
protected resources using SIF requests (RegisterUsers, UnregisterUsers, and
ListSiperianObjects). All policy decisions are external to Informatica MDM
Hub.
- 627 -
8. For non-administrator users who will interact with Informatica MDM Hub
using the Hub Console, provide them with access to the Hub Console tools
that they will need to use, as described in "Configuring User Access to ORS
Databases" on page 653. For example, data stewards typically need
access to the Merge Manager and Data Manager tools (which are described
in the Informatica MDM Hub Data Steward Guide).
If you are using external security providers instead to handle any portion of
security in your Informatica MDM Hub implementation, you must configure
them in the Hub Console, as described in "Managing Security Providers" on
page 664.
- 628 -
Service / Task "Internal- "External "Roles-based "Comprehensive
only PDP" User Centralized Centralized
on page Directory" PDP" on page PDP" on page
625 on page 626 626
625
"Assigning Resource X X X
Privileges to Roles" on
page 641
Security Providers
"Managing Security X X X
Providers" on page
664
Role Assignment
"Assigning Roles to X X
Users and User
Groups" on page 662
- 629 -
Resource Notes
Type
PACKAGE User has access to all secure packages and their columns. For
details, see "Configuring Packages" on page 151.
REMOTE User has access to all secure remote packages. Applicable only to
PACKAGE Informatica MDM Hub implementations with an Activity Manager
license.
In addition, the Hub Console allows you to protect other resources that are
accessible by SIF requests, including content metadata, match rule sets,
metadata, batch groups, validate metadata, the audit table, and the users
table.
There are certain Informatica MDM Hub resources that you might not want to
expose to external applications. For example, your Informatica MDM Hub
implementation might have mappings or packages that are used only in batch
jobs (not in SIF requests), so these could remain private.
- 630 -
Privileges
With Informatica MDM Hub internal authorization, each role is assigned one of
the following privileges.
Privilege Allows the User To....
READ View but not change data.
CREATE Create data records in the Hub Store.
UPDATE Update data records in the Hub Store.
MERGE Merge and unmerge data.
EXECUTE Execute cleanse functions (see "Using Cleanse Functions" on page
314) and batch groups (see "Running Batch Jobs Using the Batch
Group Tool" on page 512).
These privileges are not enforced when using the Hub Console, although the
settings still affect the use of Hub Console to some degree. For example, the
only packages that data stewards can see in the Merge Manager and Data
Manager tools are those packages to which they have READ privileges. In
order for data stewards to edit and save changes to data in a particular
package, they must have UPDATE and CREATE privileges to that package (and
associated columns). If they do not have UPDATE or CREATE privileges, then
any attempts to change the data in the Data Manager will fail. Similarly, a
data steward must have MERGE privileges to merge or unmerge records using
the Merge Manager. To learn more about the Merge Manager and Data
Manager tools, see the Informatica MDM Hub Data Steward Guide.
Resource Groups
A resource group can also contain other resource groups—except itself or any
resource group to which it belongs—allowing you to build a hierarchy of
- 631 -
resource groups and to simplify the management of a large collection of
resources.
- 632 -
Column Description
Resources Used to set the status of individual Informatica MDM Hub resources
(SECURE or PRIVATE). Informatica MDM Hub resources organized
in a hierarchy that shows the relationships among resources.
Global resources appear at the top of the hierarchy. For details,
see "Configuring Resources" on page 633.
Resource Used to configure resource groups. For details, see "Configuring
Groups Resource Groups" on page 635.
Configuring Resources
Use the Resources tab in the Secure Resources tool to browse and configure
Informatica MDM Hub resources.
• Click to expand the entire tree (if you have acquired a write lock).
• Click to collapse the entire tree (if you have acquired a write lock).
- 633 -
Setting the Status of a Informatica MDM Hub Resource
You can configure the resource status (SECURE or PRIVATE) for any concrete
Informatica MDM Hub resource.
Note: This status setting does not apply to resource groups (which contain
only SECURE resources) or to global resources (for example, BASE_
OBJECT.*)—only to the resources that they contain.
Filtering Resources
- 634 -
The Secure Resources tool displays the Resources Filter dialog.
4. Do the following:
• Check (select) the resource type(s) that you want to include in the
filter.
• Uncheck (clear) the resource type(s) that you want to exclude in the
filter.
5. Click OK.
The Secure Resources tool displays the filtered Resources tree.
While editing Resource Group A, the Address base object is slightly grayed.
In this example, you cannot change the check box for the Address base object
when you are editing Resource Group A. You can change the check box only
when editing Resource Group B.
- 635 -
3. Click the Resource Groups tab.
The Secure Resources tool displays the Resource Group tab.
- 636 -
The Secure Resources tool displays the Assign Resources to Resource
Group dialog.
- 637 -
To set the refresh rate for security changes:
• Set the following parameter in the cmxserver.properities file:
cmx.server.sam.cache.resources.refresh_interval
Configuring Roles
This section describes how to configure roles for your Informatica MDM Hub
implementation.
About Roles
In Informatica MDM Hub, a role represents a set of privileges to access secure
Informatica MDM Hub resources. In order for a user to view or manipulate a
secure Informatica MDM Hub resource, that user must be assigned a role that
grants them sufficient privileges to access the resource. Roles determine what
a user is authorized to access and do in Informatica MDM Hub. For more
information, see "Authorization" on page 623 and "Privileges" on page 631.
Informatica MDM Hub roles are highly granular and flexible, allowing
administrators to implement complex security safeguards according to your
organization’s unique security policies, procedures, and requirements. Some
users might be assigned to a single role with access to everything (such as an
administrator) or with explicitly-restricted privileges (such as a data
steward), while others might be assigned to multiple roles of varying
privileges.
A role can also have other roles assigned to it, thereby inheriting the access
privileges configured for those roles. Privileges are additive, meaning that,
when roles are combined, their privileges are combined as well. For example,
suppose Role A has READ privileges to an Address base object, and Role B has
CREATE and UPDATE privileges to it. If a user account is assigned Role A and
Role B, then that user account will have READ, CREATE, and UPDATE privileges
to the Address base object. A user account inherits the privileges configured
for any role to which the user account is assigned.
Resource privileges vary depending on the scope of access that is required for
users to do their jobs—ranging from broad and deep access (for example,
super-user administrators) to very narrow, focused access (for example,
- 638 -
READ privileges on one base object). It is generally recommended that you
follow the principle of least privilege—users should be assigned the least set
of privileges needed to do their work.
Because Informatica MDM Hub provides you with the ability to vary resource
privileges per role, and because resource privileges are additive, you can
define roles in a highly-granular manner for your Informatica MDM Hub
implementation. For example, you could define separate roles to provide
different access levels to human resources data (such as HRAppReadOnly,
HRAppCreateOnly, and HRAppUpdateOnly), and then combine them into
another aggregate role (such as HRAppAll). You would then assign to various
users just the role(s) that are appropriate for their job function.
Adding Roles
To add a new role:
1. Start the Roles tool. For more information, see "Starting the Roles Tool" on
page 639.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Point anywhere in the navigation pane, right-click the mouse, and choose
Add Role.
The Roles tool displays the Add Role dialog.
- 639 -
4. Specify the following information.
Field Description
Name Name of this role. Enter a unique, descriptive name.
Description Optional description of this role.
External External name (alias) of this role. For more information, see
Name "Mapping Internal Roles to External Roles" on page 641.
5. Click OK.
The Roles tool adds the new role to the roles list.
Editing Roles
To edit an existing role:
1. Start the Roles tool. For more information, see "Starting the Roles Tool" on
page 639.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Scroll the roles list and select the role that you want to edit.
4. For each property that you want to edit, click the Edit button next to it,
and specify the new value.
You can also assign and edit resource privileges for roles. For more
information, see "Assigning Resource Privileges to Roles" on page 641.
Inheriting Privileges
You can also edit the privileges for a specific role to inherit privileges from
other roles. For more information, see "Assigning Roles to Other Roles" on
page 643.
- 640 -
Deleting Roles
To delete an existing role:
1. Start the Roles tool. For more information, see "Starting the Roles Tool" on
page 639.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Scroll the roles list and select the role that you want to delete.
4. Point anywhere in the navigation pane, right-click the mouse, and choose
Delete Role.
The Roles tool prompts you to confirm deletion.
5. Click Yes.
The Roles tool removes the deleted role from the roles list.
- 641 -
4. Click the Resource Privileges tab.
The Roles tool displays the Resource Privileges tab.
Field Description
Resources Hierarchy of secure Informatica MDM Hub resources. Displays
only those Informatica MDM Hub resources whose status has
been set to SECURE in the Secure Resources tool. For more
information, see "Setting the Status of a Informatica MDM Hub
Resource" on page 634.
Privileges Privileges to assign to secure resources. For more information,
see "Privileges" on page 631.
5. Expand the Resources hierarchy to show the secure resources that you
want to configure for this role.
- 642 -
7. Click the Save button to save your changes.
The Roles tool displays any role(s) that can be assigned to the selected
role.
5. Check (select) any role that you want to assign to the selected role.
6. Uncheck (clear) any role that you want to remove from this role.
- 643 -
Generating a Report of Resource Privileges for
Roles
You can generate a report that describes only the resource privileges granted
to a given role.
5. Click Generate. The Roles tool generates the report and displays it on the
tab.
- 644 -
Clearing the Report Window
- 645 -
The Security Access Manager saves the report using the following naming
convention:
<ORS_Name>-<Role_Name>-RolePrivilegeReport.html
where:
• ORS_Name—Name of the target database.
• Role_Name—Role associated with the generated report.
The Roles tool saves the current report as an HTML file in the target
location. You can subsequently display this report using a browser.
- 646 -
• multiple users will run the Hub Console using different accounts (for
example, administrators and data stewards).
User Accounts
Users are represented in Informatica MDM Hub by user accounts, which are
defined in the master database in the Hub Store. You use the Users tool in the
Configuration workbench to define and configure user accounts for
Informatica MDM Hub users, as well as to change passwords and enable
external authentication. External applications with sufficient authorization can
also register user accounts using SIF requests, as described in the Informatica
MDM Hub Services Integration Framework Guide. A user needs to be defined
only once, even if the same user will access more than one ORS associated
with the Master Database.
A user account gains access to Informatica MDM Hub resources using the
role(s) assigned to it, inheriting the privileges configured for each role, as
described in "About Roles" on page 638.
Informatica MDM Hub allows for multiple concurrent SIF requests from the
same user account. For an external application in which granular auditing and
user tracking is not required, multiple users can use the same user account
when submitting SIF requests.
- 647 -
Starting the Users Tool
To start the Users tool:
1. In the Hub Console, connect to the master database, if you have not
already done so.
2. Expand the Configuration workbench and click Users.
The Hub Console displays the Users tool.
Configuring Users
This section describes how to configure users in the Users tool. It refers to
functionality that is available on the Users tab of the Users tool.
4. Click the button. The Users tool displays the Add User dialog.
- 648 -
5. Specify the following settings for this user.
Property Description
First name First name for this user.
Middle name Middle name for this user.
Last name Last name for this user.
User name Name of the user account for this user. Name that this
user will enter to log into the Hub Console.
Default Default database for this user. This is the database that is
database automatically selected when the user logs into Hub
Console, as described in "Starting the Hub Console" on
page 30. If you want to change this database later, see
"Configuring User Access to ORS Databases" on page 653.
Password Password for this user. If you want to change this
password later, see "Changing Password Settings for User
Accounts" on page 652.
Verify Type the password again to verify.
password
Use external One of the following settings:
authentication? • Check (select) this option to use external
authentication using a third-party security provider
instead of Informatica MDM Hub’s default
authentication. For more information, see "Managing
Security Providers" on page 664.
• Uncheck (clear) this option to use the default
Informatica MDM Hub authentication.
6. Click OK.
The Users tool adds the new user to the list of users on the Users tab.
For each user, you can update their name, their default login database, and
specify other settings—such as whether Informatica MDM Hub retains a log of
- 649 -
user logins/logouts, whether they can log into Informatica MDM Hub, and
whether they have administrator-level privileges.
Property Description
Administrator One of the following settings:
• Check (select) this option to give this user
administrative access, which allows them to have access
to all Hub Console tools and all databases.
• Uncheck (clear) this option if you do not want to grant
administrative access to this user. This is the default.
Enable One of the following settings:
• Check (select) this option to activate this user account
and allow this user to log in.
• Uncheck (clear) this option to disable this user account
and prevent this user from logging in.
- 650 -
Editing Supplemental User Information
In Informatica MDM Hub implementations that are not tied to an external user
directory (see "External User Directory" on page 625), you can use
Informatica MDM Hub to manage supplemental information for each user,
such as their e-mail address and phone numbers. Informatica MDM Hub does
not require that you provide this information, nor does Informatica MDM Hub
use this information in any special way.
Property Description
Title User’s title, such as Dr. or Ms. Click the drop-down list
and select a title.
- 651 -
Property Description
Initials User’s initials.
Suffix User’s suffix, such as MD or Jr.
Job title User’s job title.
Email User’s e-mail address.
Telephone area Area code for user’s telephone number.
code
Telephone User’s telephone number.
number
Fax area code Area code for user’s fax number.
Fax number User’s fax number.
Mobile area Area code for user’s mobile phone.
code
Mobile number User’s mobile phone number.
Login message Message that the Hub Console displays after this user
logs in.
7. Click OK.
To remove a user:
1. Start the Users tool. For more information, see "Starting the Users Tool"
on page 648.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Click the Users tab.
4. Select the user that you want to remove.
- 652 -
5. Click the button.
The Users tool displays the Change Password dialog for the selected user.
6. Specify the new password and in both the Password and Verify password
fields, if you want.
7. Do one of the following:
• Check (select) this option to use external authentication using a third-
party security provider instead of Informatica MDM Hub’s default
authentication. For more information, see "Managing Security
Providers" on page 664.
• Uncheck (clear) this option to use the default Informatica MDM Hub
authentication.
8. Click OK.
- 653 -
4. Expand each database node to see which users that can access that
database.
5. To change user assignments to a database, right-click on the database
name and choose Assign User.
The Users tool displays the Assign User to Database dialog.
6. Check (select) the names of any users that you want to assign to the
selected database.
7. Uncheck (clear) the names of any users that you want to unassign from
the selected database.
8. Click OK.
The global password policy applies to users who do not have private password
policies specified for them (as described in "Specifying Private Password
Policies for Individual Users" on page 656).
- 654 -
4. Specify the following password policy settings.
Policy Description
Password Minimum and maximum length, in characters.
Length
Password Do one of the following:
Expiry • Check (select) the Password Expires check box and
specify the number of days before the password
expires.
• Uncheck (clear) the Password Expires check box so that
the password never expires.
Login Number of grace logins and maximum number of failed
Settings logins.
Password Number of times that a password can be re-used.
History
Password Other configuration settings, such as:
Requirements • enforce case-sensitivity
• enforce password validation
• enforce a minimum number of unique characters
• password patterns
- 655 -
Specifying Private Password Policies for Individual Users
For any given user, you can specify a private password policy that overrides
the global password policy (see "Managing the Global Password Policy" on
page 654).
- 656 -
9. Click the Save button to save your changes.
To configure user names and passwords for a secured JDBC data source in the
cmxserver.properties file, use the following parameters:
databaseId.username=username
databaseId.password=encryptedPassword
For an Oracle Service connection type, the databaseId consists of the following
strings:
<service_name>-<schema_name>
- 657 -
the username and password properties would be:
MDM_Service-Test_ORS.username=weblogic
MDM_Service-Test_ORS.password=9C03B113CD8E4BBFD236C56D5FEA56EB
If you want to secure the JDBC data source that accesses the Master
Database, the databaseId is CMX_SYSTEM. In this case, the properties would
be:
CMX_SYSTEM.username=weblogic
CMX_SYSTEM.password=9C03B113CD8E4BBFD236C56D5FEA56EB
You use the Groups tab in the Users and Groups tool in the Security Access
Manager workbench to configure users groups and assign user accounts to
user groups. To use the Users and Groups tool, you must be connected to an
ORS.
- 658 -
The Hub Console displays the Users and Groups tool.
- 659 -
5. Enter a descriptive name for the user group.
6. Optionally, enter a description of the user group.
7. Click OK.
The Users and Groups tool adds the new user group to the list.
5. For each property that you want to edit, click the Edit button next to it,
and specify the new value.
- 660 -
Deleting User Groups
To delete a user group:
1. Start the Users and Groups tool. For more information, see "Starting the
Users and Groups Tool" on page 658.
2. Acquire a write lock. For more information, see "Acquiring a Write Lock"
on page 36.
3. Click the Groups tab.
4. Scroll the list of user groups and select the user group that you want to
delete.
6. Check (select) the names of any users and user groups that you want to
assign to the selected user group.
7. Uncheck (clear) the names of any users and user groups that you want to
unassign from the selected user group.
8. Click OK.
- 661 -
User Access to ORS Databases" on page 653.
5. Check (select) the names of any users that you want to assign to the
selected ORS database.
6. Uncheck (clear) the names of any users that you want to unassign from
the selected ORS database.
7. Click OK.
You can choose the way that is most expedient for your implementation.
- 662 -
4. Select the role to which you want to assign users and user groups.
6. Check (select) the names of any users and user groups that you want to
assign to the selected role.
7. Uncheck (clear) the names of any users and user groups that you want to
unassign from the selected role.
8. Click OK.
4. Select the user or user group to which you want to assign roles.
6. Check (select) the roles that you want to assign to the selected user or
user group.
7. Uncheck (clear) the roles that you want to unassign from the selected user
or user group.
- 663 -
8. Click OK.
Internal Providers
Informatica MDM Hub comes with a set of default internal security providers
(labeled Internal Provider in the Security Providers tool). You can also add
your own third-party security providers. Internal security providers cannot be
removed.
- 664 -
In the Security Providers tool, the navigation tree has the following main
nodes:
Tab Description
Provider Expand to display the provider files that have been uploaded in your
Files Informatica MDM Hub implementation. For more information, see
"Managing Provider Files" on page 665.
Providers Expand to display the list of providers that are defined in your
Informatica MDM Hub implementation. For more information, see
"Managing Security Provider Settings" on page 668.
- 665 -
• A manifest that describes one or more external security provider(s). Each
security provider has the following settings:
• Provider Name
• Provider Description
• Provider Type
• Provider Factory Class Name
• Properties for configuring the provider (a list of name-value pairs:
property names with default values)
• One or more JAR files containing the provider implementation and any
required third-party libraries.
The Security Providers tool displays a list of provider files under the Provider
Files node in the left navigation pane. You use right-click menus in the left
navigation pane of the Security Providers tool to upload, delete, and move
provider files in the Provider Files list.
- 666 -
Uploading a Provider File
The Security Provider tool prompts you to select the JAR file for this
provider.
4. Specify the JAR file, navigating the file system as needed and selecting the
JAR file that you want to upload.
5. Click Open.
- 667 -
The Security Provider tool checks the selected file to determine whether it
is a valid provider file.
If the provider name from the manifest is the same as the name of an
existing provider file, then the Security Provider tool asks you whether to
overwrite the existing provider file. Click Yes to confirm.
The Security Provider tool uploads the JAR file to the application server,
adds the provider file to the list, populates the Providers list with the
additional provider information, and refreshes the left navigation pane.
Once the file has been uploaded, the original file can be removed from the
file system, if you want. The Security Provider tool has already imported
the information and does not subsequently refer to the original file.
Note: Internal security providers that are shipped with Informatica MDM Hub
cannot be removed. For internal security providers, there is no separate
provider file under the Provider Files node.
You use right-click menus in the left navigation pane of the Security Providers
tool to move providers up and down in the Providers list.
The order of providers in the Provider list represents the order in which they
are invoked. For example, when a user attempts to log in and supplies their
- 668 -
user name and password, Informatica MDM Hub submits their login
credentials to each authentication provider in the Authentication list,
proceeding sequentially through the list. If authentication succeeds with one of
the providers in the list, then the user is deemed authenticated. If
authentication fails with all available authentication providers, then
authentication for that user fails. To learn about changing the processing
order, see "Moving a Security Provider Up in the Processing Order" on page
676 and "Moving a Security Provider Down in the Processing Order" on page
676.
- 669 -
Field Description
specify here. For more information, see "Configuring Provider
Properties" on page 670.
4. For each property that you want to edit, click the Edit button next to it,
and specify the new value.
- 670 -
5. Click the Save button to save your changes.
Custom-added Providers
You can also package custom provider classes in the JAR/ZIP file. Specify the
settings for the custom providers in properties file named
providers.properties. You must place this file within the JAR file in the META-
INF directory. These settings (that is, the name/value pairs) are then read by
the loader and translated to what is displayed in the Hub Console.
Element Description
Name
ProviderList Comma-separated list of the contained provider names.
File- Description of the package.
Description
Note that the remaining elements listed below come in groups of
five (5) which correspond to each of the names in ProviderList
(so for the remaining elements listed here, “XXX” represents one
of the names that would be specified in ProviderList).
XXX- Display name of the provider XXX.
Provider-
Name
XXX- Description of the provider XXX.
Provider-
Description
XXX- Type of the provider XXX. The allowed values are USER_
Provider- PROFILE_PROVIDER, JAAS_LOGIN_MODULE, AUTHORIZATION_
Type PROVIDER.
XXX- Implementation class of the provider (contained in the same
Provider- JAR/ZIP file).
Factory-
Class-
- 671 -
Element Description
Name
Name
XXX- Comma-separated list of name/value pairs defining provider
Provider- properties (name1=value1,…).
Properties
Note: The provider archive file (JAR/ZIP) must contain all the classes
required for the custom provider to be functional, as well as all of the required
resources. These resources are specific to your implementation.
Informatica MDM Hub supports the use of external authentication for users
through the Java Authentication and Authorization Service (JAAS). Informatica
MDM Hub provides templates for the following types of authentication
standards:
• Lightweight Directory Access Protocol (LDAP)
• Microsoft Active Directory
• Network authentication using the Kerberos protocol
- 672 -
These templates provide the settings (protocols, server names, ports, and so
on) that are required for these authentication standards. You can use these
templates to add a new login module and provide the settings you need. To
learn more about these authentication standards, see the applicable vendor
documentation.
4. Click the down arrow and select a template for the login module.
- 673 -
6. In the Properties panel, click the Edit button next to any property that
you want to edit, such as its name and description, and change the setting.
For LDAP, you can specify the following settings.
Property Description
java.naming.factory.initial Required. Java class name of the JNDI
implementation for connecting to an LDAP
server. Use the following value:
com.sun.jndi.ldap.LdapCtxFactory.
java.naming.provider.url Required. URL of the LDAP server. For example:
ldap://localhost:389/
username.prefix Optional. Tells Informatica MDM Hub how to
parse the LDAP username. An OpenLDAP user
name looks like this:
cn=myopenldapuser,dc=siperian,dc=com
where
• myopenldapuser is the user name
• siperian is the domain name
• com is the top-level domain
In this example, the username.prefix is: cn=
username.postfix Optional. User in conjunction with
username.prefix. Using the previous example,
set username.postfix to:
,dc=siperian,dc=com
Note the comma in the beginning of the string.
For Microsoft Active directory, you can specify the following settings:
Property Description
java.naming.factory.initial Required. Java class name of the JNDI
implementation for connecting to an LDAP
server. Use the following value:
com.sun.jndi.ldap.LdapCtxFactory.
java.naming.provider.url Required. URL of the LDAP server. For example:
ldap://localhost:389/
For Kerberos authentication:
- 674 -
• To set up Kerberos authentication for a user on JBoss and WebLogic
using Sun’s JVM, use Sun’s LoginModule
(com.sun.security.auth.module.Krb5LoginModule). For more
information, see the Kerberos documentation at http://java.sun.com.
• To set up Kerberos authentication for a user on WebSphere using IBM’s
JVM, you can use IBM’s LoginModule
(com.ibm.security.auth.module.Krb5LoginModule). For more
information, see the Kerberos documentation on http://www.ibm.com.
• To use either of these Kerberos implementations, you must configure
the JVM of the Informatica MDM Hub application server with
winnt\krb5.ini or
JAVA_HOME\jre\lib\security\krb5.conf.
4. In the Properties panel, click the Edit button next to any property that
you want to edit.
- 675 -
Enabling and Disabling Security Providers
1. Acquire a write lock, if you have not already done so.
2. Select the security provider that you want to enable or disable, as
described in "Selecting a Security Provider" on page 669.
3. Do one of the following:
• Check the Enabled check box to enable a disabled security provider.
• Uncheck the Enabled check box to disable a security provider.
Once disabled, the provider name appears greyed out and at the end of
the Providers list. Disabled providers cannot be moved.
- 676 -
3. In the left navigation pane, click the provider (not the last one in the list,
nor any disabled providers) that you want to move down.
4. In the left navigation pane, right-click and choose Move Provider Down.
The Security Provider tool moves the provider after the subsequent one in
the Providers list and refreshes the left navigation pane.
- 677 -
Chapter 21: Viewing Registered
Custom Code
This chapter describes how to use the User Object Registry tool to view
registered custom code.
Chapter Contents
• "About User Objects" on page 678
• "About the User Object Registry Tool" on page 678
• "Starting the User Object Registry Tool" on page 679
• "Viewing User Exits" on page 679
• "Viewing Custom Stored Procedures" on page 680
• "Viewing Custom Java Cleanse Functions" on page 681
• "Viewing Custom Button Functions" on page 682
- 678 -
Note: To view custom user code in the User Object Registry tool, you must
have registered the following types of objects:
• Custom Stored Procedures; for more information regarding stored
procedures, see "Developing Custom Stored Procedures for Batch Jobs" on
page 604
• Custom Java Cleanse Functions; for more information regarding Java
cleanse functions, see "Using Cleanse Functions" on page 314
• Custom Button Functions; for more information regarding custom buttons,
see "About Custom Buttons in the Hub Console" on page 730
Note: You do not need to pre-configure user exit procedures to view them in
the User Object Registry tool.
- 679 -
process run. User exits are triggered by the Informatica MDM Hub back-end
processes that provide a mechanism to integrate custom operations with Hub
Server processes such as POST_LOAD, POST_MERGE, POST_MATCH, and so
on. For more information, see "About User Exits" on page 708.
Note: The User Object Registry tool displays the types of pre-existing user
exits.
Informatica MDM Hub also allows you to create and run custom stored
procedures for batch jobs. For more information, see "Developing Custom
Stored Procedures for Batch Jobs" on page 604. You can also create and run
stored procedures using the SIF API (using Java, SOAP, or HTTP/XML). For
more information, see the Informatica MDM Hub Services Integration
Framework Guide.
- 680 -
How Custom Stored Procedures Are Registered
You must register a custom stored procedure with Informatica MDM Hub in
order to make it available to users in the Batch Viewer and Batch Group tools
in the Hub Console. For more information, see "Registering a Custom Stored
Procedure" on page 605.
- 681 -
How Custom Java Cleanse Functions Are Registered
Cleanse functions are configured using the Cleanse Functions tool in the Hub
Console. For more information, see "Configuring Java Libraries" on page 318
Server and client-based custom functions are visible in the User Object
Registry. For more information, see "Server-Based and Client-Based Custom
Functions" on page 732.
- 682 -
How Custom Button Functions Are Registered
To add a custom button to the Hub Console in your Informatica MDM Hub
implementation, complete the following tasks:
1. Determine the details of the external service that you want to invoke, such
as the format and parameters for request and response messages.
2. Write and package the business logic that the custom button will execute,
as described in "Writing a Custom Function" on page 732.
3. Deploy the package so that it appears in the applicable tool(s) in the Hub
Console, as described in "Deploying Custom Buttons" on page 735.
- 683 -
Chapter 22: Auditing Informatica
MDM Hub Services and Events
This chapter describes how to set up auditing and debugging in the Hub
Console.
Chapter Contents
• "About Integration Auditing" on page 684
• "Starting the Audit Manager" on page 686
• "Auditing SIF API Requests" on page 688
• "Auditing Message Queues" on page 689
• "Auditing Errors" on page 690
• "Using the Audit Log" on page 691
Auditable Events
Integration with external applications often involves complexity. Multiple
applications interact with each other, exchange data synchronously or
asynchronously, use data transformations back and forth, and engage various
business rules to execute business processes across applications.
- 684 -
• Informatica MDM Hub sends a message (using JMS) to a message queue
for the purpose of distributing data changes to other systems. For more
information, see "Configuring the Publish Process" on page 449.
For more information, see "Starting the Audit Manager" on page 686.
- 685 -
credentials, that information will not be captured in the C_REPOS_AUDIT
table. Auditing begins only after authentication succeeds.
- 686 -
Type Description
API Request invocations made by external applications using the
Requests Services Integration Framework (SIF) Software Development Kit
(SDK).
Message Message queues used for message triggers. For more information,
Queues see "Configuring the Publish Process" on page 449.
Note: Message queues are defined at the CMX_SYSTEM level. These
settings apply only to messages for this Operational Reference
Store (ORS).
Systems to Audit
For each type of item to audit, the Audit Manager displays the list of systems
that can be audited, along with the SIF requests that are associated with that
system.
System Description
No System Services that are not—or not necessarily—associated with a
specific system (such as merge operations).
Admin Services that are associated with the Admin system.
Defined Services that are associated with predefined source systems.
Source For more information, see "About the Databases Tool" on page
Systems 54.
Note: The same API request or message queue can appear in multiple source
systems if, for example, its use is optional on one of those source systems.
Audit Properties
Note: A write lock is not required to configure auditing.
When you select an item to audit, the Audit Manager displays properties in the
properties pane with the following configurable settings.
Field Description
System Name of the selected system. Read-only.
Name
Description Description of the selected system. Read-only.
API List of API requests that can be audited.
Request
Message List of message queues that can be audited.
Queue
Enable By default, auditing is not enabled.
Audit? • Select (check) to enable auditing for the item.
• Clear (uncheck) to disable auditing for the item.
Include This check box is available only if auditing is enabled for this item.
XML? By default, capturing XML in the log is not included. For more
information, see "Capturing XML for Requests and Responses" on
page 685.
• Check (select) to include XML in the audit log for this item.
• Uncheck (clear) to exclude XML from the audit log for this
item.
- 687 -
Field Description
Note: Passwords are never stored in the audit log. If a password
exists in the XML stream (whether encrypted or not), Informatica
MDM Hub replaces the password with asterisks:
...<get>
<username>admin</username>
<password>
<encrypted>false</encrypted>
<password>******</password>
</password>
...
Important: Selecting this option can cause the audit log file to
grow very large rapidly. For more information, see "Periodically
Purging the Audit Log" on page 696.
For the Enable Audit? and Include XML? check boxes, you can use the following
buttons.
For more information regarding the SIF API requests, see Informatica MDM
Hub Services Integration Framework Guide.
- 688 -
3. For each SIF request that you want to audit, select (check) the Enable
Audit check box.
4. If auditing is enabled for a particular API request and you also want to
include XML associated with that API request in the audit log, then select
(check) the Include XML check box.
- 689 -
3. For each message queue that you want to audit, select (check) the Enable
Audit check box.
4. If auditing is enabled for a particular message queue and you also want to
include XML associated with that message queue in the audit log, then
select (check) the Include XML check box.
Auditing Errors
You can capture error information for any SIF request invocation that triggers
the error mechanism in the Web service—such as syntax errors, run-time
errors, and so on. You can enable auditing for all errors associated with SIF
requests.
Auditing errors is a feature that you enable globally. Even when auditing is not
currently enabled for a particular SIF request, if an error occurs during that
SIF request invocation, then the event is captured in the audit log.
- 690 -
3. Do one of the following:
• Select (check) the Enable Audit check box to audit errors.
• Clear (uncheck) the Enable Audit check box to stop auditing errors.
4. If you select Enable Audit and you also want to include XML associated
with errors in the audit log, then select (check) the Include XML check
box.
Note: If you only select Enable Audit, Informatica MDM Hub provides the
associated audit information in C_REPOS_AUDIT.
If you also select Include XML, Informatica MDM Hub includes an
additional column in C_REPOS_AUDIT named DATA_XML which includes
detail log data for audit.
If you select both check boxes, when you run an Insert, Update, or Delete
job in the Data Manager, or run the associated batch job, Informatica MDM
Hub includes the audit data in DATA_XML of C_REPOS_AUDIT.
Note: The SIF Audit request allows an external application to insert new
records in the C_REPOS_AUDIT table. You would use this request to report
- 691 -
activity involving a record(s) in Informatica MDM Hub, that is at a higher
level, or has more information that can be recorded by the Hub. For example,
audit an update to a complex object before transforming and decomposing it
to Hub objects. For more information, see the Informatica MDM Hub Services
Integration Framework Guide.
- 692 -
Name Oracle Type DB2 Type Description
so on. If an error occurs, the
request XML is always put in this
column to ensure its capture in
case auditing was not enabled
for the SIF request that was
invoked. Populated only if the
Include XML option is enabled
(checked).
ROWID_ CHAR(14) CHARACTER(14) Reference to the ROWID_AUDIT
AUDIT_ of the related previous entry.
PREVIOUS For example, links a response
entry to its corresponding
request entry.
INTERACTION_ NUMBER(19) BIGINT(8) Interaction ID. May be NULL
ID since INTERACTION_ID is
optional.
USERNAME VARCHAR2(50) VARCHAR(50) User that invoked the SIF
request. Null for message
queues.
FROM_SYSTEM VARCHAR2(50) VARCHAR(50) Source system for a SIF
request, or Admin for message
queues.
TO_SYSTEM VARCHAR2(50) VARCHAR(50) System to which the audited
event is related.
For example, API Requests to
Hub set this to “Admin” and the
responses are the system or
null if not known (and vice-
versa for Responses).
Note: Activity Manager Actions
set this value.
TABLE_NAME VARCHAR2(100) VARCHAR(100) Table in the Hub Store that is
associated with this audited
event.
CONTEXT VARCHAR2(255) VARCHAR(255) Metadata. For example,
pkeySource
This is null for audits from Hub,
but may have values for Activity
Manager and audits done
through the SIF API.
- 693 -
If available in the data management tool you use to view the log file, you can
focus your viewing by filtering entries—by audit level (view only debug-level
or info-level entries), by time (view entries within the past hour), by operation
success / failure (show error entries only), and so on.
- 694 -
Here is an example C_REPOS_AUDIT with audit log entries that includes the
XML column. For this example, both Enable Audit and Include XML check
boxes were enabled.
- 695 -
Periodically Purging the Audit Log
The audit log table can grow very large rapidly, particularly when capturing
XML request and response information (when the Include XML option is
enabled). Using tools provided by your database management system,
consider setting up a scheduled job that periodically deletes records matching
a particular filter (such as entries created more than 60 minutes ago).
- 696 -
Part 6: Appendixes
Part 6: Appendixes
Contents
• "Configuring International Data Support" on page 698
• "Backing Up and Restoring Informatica MDM Hub" on page 706
• "Configuring User Exits" on page 708
• "Viewing Configuration Details" on page 715
• "Implementing Custom Buttons in Hub Console Tools" on page 730
• "Configuring Access to Hub Console Tools" on page 737
• "Row-level Locking" on page 740
- 697 -
Appendix A: Configuring
International Data Support
This topic explains how to configure character sets in a Informatica MDM Hub
implementation. The database needs to support the character set you want to
use, the terminal must be configured to support the character set you want to
use, and the NLS_LANG environment variable must include the Oracle name
for the character set used by your client terminal.
Appendix Contents
• "Configuring Unicode in Informatica MDM Hub" on page 698
• "Configuring the ANSI Code Page (Windows Only)" on page 703
• "Configuring NLS_LANG" on page 704
Notes:
• The NLS_LANG setting should match the database character set.
- 698 -
• The language_territory portion of the NLS_LANG setting (represented
as “AMERICA_AMERICA” in the above example) is locale-specific and might
not be suitable for all Informatica MDM Hub implementations. For
example, a Japanese implementation might need to use the following
setting instead:
NLS_LANG=JAPANESE_JAPAN.AL32UTF8
• If you use AL32UTF8 (or even UTF8) as the database character set,
then it is highly recommended that you set NLS_LENGTH_SEMANTICS
to CHAR (in the Oracle init.ora file) when you instantiate the database.
Doing so forces Oracle to default to CHAR (not BYTE) for variable
length definitions. The NLS_LENGTH_SEMANTICS setting affects all
character-related variable types: VARCHAR, VARCHAR2, and CHAR.
3. Ensure that the Regional Font Settings are correctly configured on the
client. For East Asian data, be sure to install East Asian fonts.
4. When editing data, the regional font settings should match the language
being used.
5. If you are using a multi-byte character set in your Oracle database, you
must change the following setting in the REPOS_DB_RELEASE table to zero
(0):
column_length_in_bytes_ind = 0
By default, this setting is one (1), which means that column lengths are
declared as byte values. Changing this to zero (0) means that column
lengths are declared as CHAR values in support of Unicode values.
Configuring Populations
By default, Informatica MDM Hub supports the population for the United States
(provides a usa.ysp file in the default installation). If your implementation
needs to use a population other than the US population, then additional
analysis of the data is required.
• If the data is exclusively from a different country, and Informatica
provides a population for that country, then use that population. Contact
Informatica Support to obtain the population.ysp file that is appropriate for
your implementation, along with instructions to enable the population.
• If the data is mostly from one country with very small amounts of mixed
data from one or more other populations, consider using the majority
population. Contact Informatica Support to obtain the population.ysp file
for the majority population, along with any instructions.
- 699 -
• If large quantities of data from different countries are mixed, consider
whether it is meaningful to match across such a disparate set of data. If
so, then consider using the “international” population. Contact Informatica
Support to obtain the appropriate population.ysp file and instructions to
enable the population.
• For all other situations, contact Informatica Support.
For example:
C:\<infamdm_install_dir>\hub\cleanse\resources\match
Unix
<infamdm_install_dir>/hub/cleanse/
Note: Informatica ships the usa.ysp file by default. If you need to use the
population set for a different country, contact Informatica Support to
obtain the population.ysp file that is appropriate for your implementation,
along with instructions to enable the population.
This setting helps with the processing of UTF8 characters during match,
ensuring that all data is represented in UTF16 (although its representation in
the database is still UTF8).
- 700 -
Using Multiple Populations Within a Single Base Object
Informatica MDM Hub provides you with the ability to use multiple populations
within a single base object. This is useful if data in a base object comes from
different populations—for example, 70% of the records from the United States
and 30% from China. Populations can vary on a record-by-record basis.
For example:
C:\<infamdm_install_dir>\hub\cleanse\resources\match
Unix
<infamdm_install_dir>/hub/cleanse/
4. Restart the application server.
5. In the Schema Manager, add a column to the base object that will contain
the population to use for each record.
This must be a VARCHAR column with the physical name of SIP_POP.
Note: The width of the VARCHAR column must fit the largest population
name in use. A width of 30 is probably sufficient for most
implementations.
6. Configure the match column as an exact match column with the name of
SIP_POP, according to the instructions in "Configuring Match Columns" on
page 387.
- 701 -
7. For each record in the base object that will use a non-default population,
provide (in the SIP_POP column) the name of the population to use
instead.
• You can specify values for the SIP_POP column in any manner of
ways—adding the data in the landing tables, using cleanse functions
that calculate the values during the stage process, invoking SIF
requests from external applications—even manually editing the cells
using the Data Manager tool. The only requirement is that the SIP_POP
cells must contain this data for all non-default populations just prior to
executing the Generate Match Tokens process.
• The data in the SIP_POP column can be in any case (upper, lower, or
mixed) because all alphabetic characters will be converted to
lowercase in the match key table. For example, Us, US, and us are all
valid values for this column.
• Invalid values in this column will be processed using the default
population. Invalid values include NULLs, empty strings, and any string
that does not match a population name as defined in c_repos_ssa_
population.population_name.
8. Execute the Generate Match Tokens process on this base object to update
the match key table.
9. Execute the match process on this base object.
Note: The match process compares only records that share the same
population. For example, it will compare Chinese records with Chinese
records, and American records with American records. Any resulting
match pairs will be between records that share the same population.
- 702 -
Hub Console
In the Hub Console, menus, warnings, and so on are in English. Current
Informatica MDM Hub UTF support applies only to business data—not
metadata or the interface. The Hub Console will have UTF8 support in a future
release.
You can configure the system locale settings (which define settings for the
system language) to use UTF-8 by completing the following steps:
1. Run the following command:
locale -a
2. Determine whether you can find a locale for your language with a name
ending in .utf8.
localedef -f UTF-8 -i en_US en_US.utf8
3. Once you know whether you have a locale that allows you to use UTF-8,
instruct the UNIX system to use that locale.
Export LC_ALL="en_US.utf8"
export LANG="en_US.utf8"
export LANGUAGE="en_US.utf8"
- 703 -
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Nls\CodePage\ACP
Note: There are many registry entries with very similar names, so be sure to
look at the right place in the registry.
Note: On Windows XP systems, you might need to install support for non-
Western languages.
Configuring NLS_LANG
To specify the locale behavior of your client Oracle software, you need to set
your NLS_LANG setting, which specifies the language, territory, and the
character set of your client. This section describes several ways in which to
configure the NLS_LANG setting.
where:
Setting Description
LANGUAGE Specifies the language used for Oracle messages, as well as the
names of days and months.
TERRITORY Specifies monetary and numeric formats, as well as territory
and conventions for calculating week and day numbers.
CHARACTERSET Controls the character set used by the client application, or it
matches your Windows code page, or it is set to UTF8 for a
Unicode application.
Note: The character set defined with the NLS_LANG parameter does not change
your client's character set. Instead, it is used to let Oracle know which
character set you are using on the client side so that Oracle can perform the
proper conversion. The character set part of the NLS_LANG parameter is never
inherited from the server.
- 704 -
Configuring NLS_LANG in the Windows Registry
On Windows systems, you should make sure that you have set an NLS_LANG
registry subkey for each of your Oracle Homes:
You can modify this subkey using the Windows Registry Editor:
1. From the Start menu, choose Run...
2. At the command prompt, type regedit, and then click OK.
3. Edit the following registry entry:
For Oracle 10g:
HKEY_LOCAL_MACHINE\SOFTWARE\ORACLE\KEY_<oracle_home_name>
When starting an Oracle tool (such as sqlplusw), the tool will read the contents
of the oracle.key file located in the same directory to determine which
registry tree will be used (therefore, which NLS_LANG subkey will be used).
- 705 -
Appendix B: Backing Up and
Restoring Informatica MDM Hub
This appendix explains how to back up and restore a Informatica MDM Hub
implementation.
Appendix Contents
• "Backing Up Informatica MDM Hub" on page 706
• "Backup and Recovery Strategies for Informatica MDM Hub" on page 706
Non-logging operations (such as CTAS, Direct Path SQL Load, and Direct Insert)
are occasionally performed on permanent Hub tables to speed-up batch
processes. These operations are not recorded in the redo logs and, as such,
are not generally recoverable. However, recovery is possible if a backup is
made immediately after the operations are completed.
Backup and recovery strategies are dependent on the value of the GLOBAL_
NOLOGGING_IND column (in the C_REPOS_DB_RELEASE table), which turns non-
logging operations on or off.
- 706 -
backup and recovery strategies: backup and recovery with non-logging
operations, and backup and recovery without non-logging operations.
To recover changes that the non-logging operations make, you must perform
an immediate back-up procedure.
3. Use the following command to disable index creation with the non-logging
option:
Run sql:
update c_repos_table set NOLOGGING_IND = 0;
COMMIT;
4. Make sure that the database is running in the archive log mode.
5. Perform a database backup.
6. If recovery is needed, apply redo logs on the backup.
- 707 -
Appendix C: Configuring User Exits
Appendix Contents
• "About User Exits" on page 708
• "Types of User Exits" on page 708
Note: The POST_LANDING, PRE_STAGE, and POST_STAGE user exits are only
called from the batch Stage process. For more information, see "Stage Jobs"
on page 556.
A user exit must perform its own transaction handling. COMMITs / ROLLBACKs
must be explicitly issued for any data manipulation operation(s) in a user exit,
or in stored procedures called from user exits. However, this is not true for
the SIF API requests (for example, Merge, Unmerge, and so on). Transactions
for the API requests are handled by Java code. Any COMMITs / ROLLBACKs in
such a case may cause a Java distributed transaction error.
Note: For Oracle databases, all user exit procedures are located in the cmxue
package.
- 708 -
User Exit Description
Name
POST_ Data in a Landing table can be refined using this user exit after
LANDING the Landing table has been populated using an ETL process. For
more information, see "POST_LANDING User Exit" on page 709.
PRE_STAGE Called before loading the data into a Staging table. For more
information, see "PRE_STAGE User Exit" on page 710.
POST_STAGE Called after a Staging table has been populated. For more
information, see "POST_STAGE User Exit" on page 710.
POST_LOAD Called after a Load batch job and after a Put API call. For more
information, see "POST_LOAD User Exit" on page 711.
PRE_MATCH Called before a Match batch job.
POST_MATCH Called after a Match batch job. For more information, see
"POST_MATCH User Exit" on page 712.
PRE_USER_ Called just before records to be merged are assigned to a user.
MERGE_ For more information, see "PRE_USER_MERGE_ASSIGNMENT"
ASSIGNMENT on page 714.
POST_MERGE Called after a Merge or a Multi-Merge batch job and after a
Merge API call. For more information, see "POST_MERGE User
Exit" on page 713.
POST_ Called after a Unmerge API call. For more information, see
UNMERGE "POST_UNMERGE User Exit" on page 713.
Use a POST_LANDING user exit for custom work on the landing table prior to
delta detection. For example:
• Hard delete detection
• Replace control characters with printable characters
• Perform any special pre-cleansing processes on Addresses
POST_LANDING Parameters
- 709 -
Parameter Description
Name
IN_ROWID_ Job id for the Stage job, as registered in C_REPOS_JOB_
JOB CONTROL.
IN_ Source table for the Stage job
LANDING_
TABLE_
NAME
IN_ Target table for the Stage job
STAGING_
TABLE_
NAME
IN_PRL_ Previous Landing table name; that is, the copy of the source
TABLE_ data mapped to the staging table from the previous time the
NAME Stage job ran
OUT_ Error message.
ERROR_
MESSAGE
OUT_ Return code.
RETURN_
CODE
Use a PRE_STAGE user exit for any special handling of delta processes. For
example, use a PRE_STAGE user exit to check delta volumes and determine
whether they exceed pre-defined allowable delta volume limits (for example,
“stop process if source system is System A and the number of deltas is
greater than 500,000”).
PRE_STAGE Parameters
Parameter Name Description
IN_ROWID_JOB Job id for the Stage job, as registered in C_REPOS_JOB_
CONTROL.
IN_LANDING_ Source table for the Stage job.
TABLE_NAME
IN_STAGING_ Target table for the Stage job.
TABLE_NAME
IN_DLT_TABLE_ Delta table name; that is, the table containing the
NAME records identified as deltas.
OUT_ERROR_ Error message.
MESSAGE
OUT_RETURN_ Return code.
CODE
Use a POST_STAGE user exit for any special processing at the end of a Stage
job.For example, use a POST_STAGE user exit for special handling of rejected
records from the Stage job (for example, to automatically delete rejects for
known, non-critical conditions).
- 710 -
POST_STAGE Parameters
Parameter Description
Name
IN_ROWID_ Job id for the Stage job, as registered in c_repos_job_control.
JOB
IN_ Source table for the Stage job
LANDING_
TABLE_
NAME
IN_ Target table for the Stage job.
STAGING_
TABLE_
NAME
IN_PRL_ Previous Landing table name; that is, the copy of the source
TABLE_ data mapped to the staging table from the previous time the
NAME Stage job ran.
OUT_ Error message.
ERROR_
MESSAGE
OUT_ Return code.
RETURN_
CODE
Use a POST_LOAD user exit after an update or after an insert from Load.
For the Load process, the IN_ACTION_TABLE has the name of the work table
containing the ROWID_OBJECT values to be inserted/updated.
POST_LOAD Parameters
- 711 -
Parameter Description
Name
IN_ Job id for the Load job, as registered in c_repos_job_control
ROWID_ (Blank for the PUT).
JOB
IN_ Name of the target table (Base Object / Relationship Table) for the
TABLE_ Load job.
NAME
IN_ Name of the source table for the Load job.
STAGE_
TABLE
IN_ For the Load job, this is the name of the table containing the rows
ACTION_ to be inserted or updated (staging_table_name_TINS for inserts,
TABLE staging_table_name_TOPT for updates).
OUT_ Error message.
ERROR_
MESSAGE
OUT_ Return code.
RETURN_
CODE
Use a POST_MATCH user exit for custom work on the match table.
POST_MATCH Parameters
Parameter Name Description
IN_ROWID_JOB Job id for the Match job, as registered in c_repos_job_
control
IN_TABLE_NAME Base Object that the Match job is running on.
IN_MATCH_SET_ Match ruleset.
NAME
OUT_ERROR_ Error message.
MESSAGE
- 712 -
Parameter Name Description
OUT_RETURN_CODE Return code.
Use a POST_MERGE user exit to perform custom work after the Merge
process.
For example, use a POST_MERGE user exit to automatically match and merge
child records affected by the match and merge of a parent record.
POST_MERGE Parameters
Parameter Name Description
IN_ROWID_JOB Job id for the Merge job, as registered in c_repos_
job_control.
IN_TABLE_NAME Base Object that the Merge job is running on.
IN_ROWID_OBJECT_ Bulk merge–action table.
TABLE On-line merge–in line view.
OUT_ERROR_MESSAGE Error message.
OUT_RETURN_CODE Return code.
Use a POST_UNMERGE user exit for custom work after the Unmerge process.
- 713 -
POST_UNMERGE Parameters
Parameter Name Description
IN_ROWID_JOB Job id for the Unmerge transaction, as registered in c_
repos_job_control.
IN_TABLE_NAME Base Object that the Unmerge job is running on.
IN_ROWID_ Re-instated rowid_object.
OBJECT
OUT_ERROR_ Error message.
MESSAGE
OUT_RETURN_ Return code.
CODE
Use this user exit to override or extend user assignment lists. This user exit
procedure runs before the user merge assignment is updated. Note that user
assignment lists are stored in C_REPOS_USER_MERGE_ASSIGNMENTS.
- 714 -
Appendix D: Viewing Configuration
Details
This appendix explains how to use the Enterprise Manager tool in the Hub
Console to configure and view the various properties, history, and database
log information in a Informatica MDM Hub implementation.
Appendix Contents
• "About the Enterprise Manager" on page 715
• "Starting the Enterprise Manager" on page 715
• "Viewing Enterprise Manager Properties" on page 716
• "Viewing Version History" on page 723
• "Using ORS Database Logs" on page 724
launch button.
The Hub Console displays the Enterprise Manager tool.
- 715 -
Viewing Enterprise Manager Properties
This section explains how to choose the different servers or databases to
view, and lists the properties that Enterprise Manager displays for the Hub
server, cleanse server, and Master Database. In addition, the Enterprise
Manager displays version history for each component.
In the Enterprise Manager screen, choose the hub component tab for the type
of information you want to view. The following options are available:
• Hub Servers
• Cleanse Servers
• Master database
• ORS databases
• Environment Report
- 716 -
Hub Server Properties
When you choose the Hub Server tab, Enterprise Manager displays the Hub
Server properties. For additional information regarding these properties, refer
to the cmxserver.properties file.
In addition, to view more information about each property, slide your cursor
or mouse over the specific property.
The following table describes Hub Server properties that Enterprise Manager
can display in the Properties tab. These properties are found in the
cmxserver.properties file (in the hub server installation directory), and are not
configurable.
Property Name Explanation Property
Installation Installation directory of the cmx.home= C:/<infamdm_install_dir>/hub/server
directory Hub Server
Master Type of Master Database cmx.server.masterdatabase.type=ORACLE
database type
Application Type of application server: cmx.appserver.type=<application_server_name>
server type JBoss, WebSphere, WebLogic
Application Optional property used to cmx.appserver.hostname=Clustername
server deploy MRM into the EJB
hostname cluster.
RMI port Application server port cmx.appserver.rmi.port=<port_#>
(depends on the appserver
type)
default settings: 2809 for
Websphere, 1099 for JBoss,
7001 for WebLogic
- 717 -
Property Name Explanation Property
Naming Naming protocol for the cmx.appserver.naming.protocol=Jnp
protocol application server type
iiop for Websphere, jnp for
JBoss, t3 for WebLogic
Initial heap Initial heap size for Java jnlp.initial-heap-size=128m
size for Java
web start JVM
Maximum heap Maximum heap size for Java jnlp.max-heap-size=512m
size for Java web start JVM
web start JVM
Refresh Refresh interval for SAM cmx.server.sam.cache.resources.refresh_
interval for resources interval=5
SAM resources Properties are specific to the
in clock ticks cmx.server.sam.cache.user_profile.refresh_
Security Access Manager interval=1
component and used to
manage cached resources for cmx.server.clock.tick_interval=60000
user profiles.
Refresh Refresh interval for SAM user cmx.server.provider.userprofile.cacheable=false
interval for profiles cmx.server.provider.userprofile.expiration=60000
SAM user
profiles in clock cmx.server.provider.userprofile.lifespan=60000
ticks
Lookup Number of entries that will be sip.lookup.dropdown.limit=100
dropdown limit populated in a dropdown menu
in the Data Manager and
Merge Manager tools.
There is no minimum or
maximum limit for this value.
Java runtime Sun Microsystems Inc.
environment
vendor
Java runtime 1.5.0_15
environment
version
- 718 -
The following table describes Cleanse Server properties that Enterprise
Manager can display in the Properties tab. These properties are found in the
cmxcleanse.properties file.
- 719 -
Property Explanation Property
Name
Number of records per match cmx.server.match.max_records_per_ranger_node=300
ranger node (limits memory
use)
Number of threads used cmx.server.cleanse.num_of_threads=1
during cleaning activities
Address Address Doctor cleanse cleanse.library.addressDoctor.property.AddressDoctor.UnlockCo
Doctor library unlock code
properties
Address Doctor cleanse cleanse.library.addressDoctor.property.AddressDoctor.Database
library database path
Address Doctor optimization cleanse.library.addressDoctor.property.AddressDoctor.Optimiza
Address Doctor memory cleanse.library.addressDoctor.property.AddressDoctor.MemoryMB
setting
Address Doctor correction cleanse.library.addressDoctor.property.AddressDoctor.Correcti
type
Address Doctor certified cleanse.library.addressDoctor.property.AddressDoctor.
preload part
Address Doctor certified cleanse.library.addressDoctor.property.AddressDoctor.PreLoad.
preload full
Address Doctor correction cleanse.library.addressDoctor.property.AddressDoctor.PreLoad.
preload part
Address Doctor correction cleanse.library.addressDoctor.property.AddressDoctor.PreLoad.
preload full
Trillium Trillium cleanse library config cleanse.library.trilliumDir.property.config.file.1=C:
properties file 1 default_config_Global.txt
Trillium cleanse library config cleanse.library.trilliumDir.property.config.file.2=C:
file 2 default_config_US_detail.txt
Trillium cleanse library config cleanse.library.trilliumDir.property.config.file.3=C:/<infamd
file 3 default_config_US_summary.txt
- 720 -
The top panel contains a list of ORS databases that are registered with the
Master Database. The bottom panel displays the properties and the version
history of the ORS database that is selected in the top panel. Properties of ORS
include database vendor and version, as well as information from the C_
REPOS_DB_RELEASE table. Version history is also kept in the C_REPOS_DB_
VERSION table.
Note: Enterprise Manager displays only ORS databases that are valid for the
current version of Informatica MDM Hub. If Enterprise Manager cannot obtain
database information for an ORS database (for example, if the ORS database
requires upgrading to the current Informatica MDM Hub version), then
Enterprise Manager displays a message explaining why the ORS database(s) is
not included in the list.
C_REPOS_DB_RELEASE Table
- 721 -
Field Name Description
0 = debug is not enabled
1 = debug is enabled
DEBUG_LEVEL_ Debug level to process (standard 5 from DEBUG to FATAL).
NUMBER Note that the default “DEBUG” is not used by standard
DEBUG_PRINT procedure.
DEBUG_LOG_ Size of database log file (in MB); default is 5.
FILE_SIZE
DEBUG_LOG_ Number of log files used for log rolling; default is 5.
FILE_NUMBER
TNSNAME TNS name of the ORS database.
CONNECTION_ Port on which the ORS database listens.
PORT
ORACLE_SID Oracle database identifier.
DATABASE_ Host on which the database is installed.
HOST
INTER_SYSTEM_ Delta-detection value, in seconds, which determines if the
TIME_DELTA_ incoming data is in the future.
SEC
COLUMN_ Flag that the SQLLoader uses to determine if the database it
LENGTH_IN_ is loading into is a UTF-8 database.
BYTES_IND A default value of 1 means that the database is UTF-8.
LOAD_TEMPLATE
MTIP_ Flag that indicates that the MTIP views will be regenerated
REGENERATION_ before the match/merge process.
REQUIRED_IND The default value of 0 (zero) means that views will not be
regenerated.
GLOBAL_ This is used when tables are created to enable logging for
NOLOGGING_ DB recovery. The default of 1 means no logging.
IND
Environment Report
When you choose the Environment Report tab, Enterprise Manager displays a
summary of the properties of all the other choices, along with any associated
error messages.
- 722 -
Saving the Hub Environment Report
- 723 -
Before you can choose servers or databases to view, you must first start
Enterprise Manager. See "Starting the Enterprise Manager" on page 715.
Each entry_process is the name of the entry stored procedure (that is, not the
name of the actual batch job or SIF API request). Here is a sample log:
- 724 -
06-JAN-2009 10:12:53.337[DEBUG][sid:103][Init_debug_vars:Init_debug_vars..........
CMXLOG.173] CMXLOG initializtion; 28 module records read.
06-JAN-2009 10:12:53.337[DEBUG][sid:103][Task Assignment
Daemon:Application.............. CMXUE.304] Start of cmxue.assign_tasks
06-JAN-2009 10:12:53.337[INFO ][sid:103][Task Assignment Daemon:Task Assignment
Daemon... CMXTASK.237] Start of Task Assignment Daemon
For a more complete sample database log file, see "Sample Database Log
File" on page 729.
Enterprise Manager includes a set of log levels to aid in the debugging and
information retrieval process. The possible database log levels are:
Name ORS Metadata Table Description
Value
ALL 500 Logs all associated information. Default log
level.
DEBUG 400 Used for debugging (default).
INFO 300 Database log information.
WARNING 200 Warning messages.
ERROR 100 Error messages.
C_REPOS_LOG_MODULE Table
When an ORS database is installed, the Hub adds a row to the C_REPOS_LOG_
MODULE table for each of the entry processes which may run stored
procedures. These module names could be batch jobs or SIF API requests. By
default, the log level for these modules is DEBUG (400).
The Hub stores the ORS debug file information in the C_REPOS_DB_RELEASE table.
For more information, see "C_REPOS_DB_RELEASE Table" on page 721.
- 725 -
Procedures for Appending Data to Database Log Files
The following PL/SQL procedures are used by the ORS CMSMIG package, and
various setup and migration scripts/methods for appending information to the
database log file.
Note: These are external stored procedures for backward compatibility for
DEBUG_PRINT functionality only.
Procedure Parameters Description
CMXLB.DEBUG_ in_debug_text, Basic debug procedure used by the ORS
PRINT in_offset database
CMXMIG.DEBUG_ in_debug_text Used only by procedures of CMXMIG
PRINT in_debug_ package
level
in_debug_file
DEBUG_PRINT in_debug_text Standalone procedure used by setup,
migration scripts, and methods of
CMX_TABLE object type
When a database log file reaches the maximum size (as defined by DEBUG_
LOG_FILE_SIZE in the C_REPOS_DB_RELEASE table), the Enterprise Manager
performs a log rolling procedure to archive the existing log information and to
prevent log files from being overridden with new database information:
1. The following settings are defined:
• Maximum file size (MaxFileSize)
• Maximum number of files (MaxBackupIndex)
• Name of the log file (debug.log)
2. The Hub logger (Log4J) appends the various database messages to the
debug.log file.
3. When the debug.log file size exceeds the MaxFileSize, the Hub activates
the log rolling procedure:
1. The active debug.log file is renamed to <filename>.hold.
2. For every file named <filename>.(n), the file is renamed to
<filename>.(n+1).
- 726 -
Note: The Hub also creates a log.rolling file prior to executing a rollover. If
your log files are not rolling as expected, check your log file directory and
remove the log.rolling file; as a result, the log rolling can continue.
4. Check the Debug Enabled check box in the top pane of the Database
Log Configuration screen.
5. Enter the debug file path, debug file name, and debug file level.
- 727 -
6. Enter the database log file size.
7. Enter the number of database log files.
8. Click Save.
9. Set the associated application server settings for debugging:
IBM WebSphere:
1. Open the following file for editing: log4j.xml
This file is in <infamdm_install_dir>\hub_5910\server\conf.
2. Change default value to DEBUG:
<category name="com.delos">
<priority value="DEBUG"/>
</category>
<category name="com.siperian">
<priority value="DEBUG"/>
</category>
<category name="siperian.performance">
<priority value="INFO"/>
</category>
<category name="com.siperian">
<priority value="DEBUG"/>
</category>
<category name="siperian.performance">
<priority value="INFO"/>
</category>
JBoss:
1. Open the following file for editing: jboss-log4j.xml
This file is in jboss-5.1.0.GA_oracle\server\default\conf
2. Change default value to DEBUG:
<category name="com.delos">
<priority value="DEBUG"/>
</category>
<category name="com.siperian">
<priority value="DEBUG"/>
</category>
<category name="siperian.performance">
- 728 -
<priority value="INFO"/>
</category>
- 729 -
Appendix E: Implementing Custom
Buttons in Hub Console Tools
This chapter explains how, in a Informatica MDM Hub implementation, you can
add custom buttons to tools in the Hub Console that allow you to invoke
external services on demand.
Appendix Contents
• "About Custom Buttons in the Hub Console" on page 730
• "Adding Custom Buttons" on page 731
Custom buttons can give users the ability to invoke a particular external
service (such as retrieving data or computing results), perform a specialized
operation (such as launching a workflow), and other tasks. Custom buttons
can be designed to access data services by a wide range of service providers,
including—but not limited to—enterprise applications (such as CRM or ERP
applications), external service providers (such as foreign exchange
calculators, publishers of financial market indexes, or government agencies),
and even Informatica MDM Hub itself (for more information, see the
Informatica MDM Hub Services Integration Framework Guide).
For example, you could add a custom button that invokes a specialized
cleanse function, offered as a Web service by a vendor, that cleanses data in
the customer record that is currently selected in the Merge Manager screen.
When the user clicks the button, the underlying code would capture the
relevant data from the selected record, create a request (possibly including
authentication information) in the format expected by the Web service, and
then submit that request to the Web service for processing. When the results
are returned, the Hub displays the information in a separate Swing dialog (if
you created one and if you implemented this as a client custom function) with
the customer rowid_object from Informatica MDM Hub.
Custom buttons are not installed by default, nor are they required for every
Informatica MDM Hub implementation. For each custom button you need to
implement a Java interface, package the implementation in a JAR file, and
- 730 -
deploy it by running a command-line utility. To control the appearance of the
custom button in the Hub Console, you can supply either text or an icon
graphic in any Swing-compatible graphic format (such as JPG, PNG, or GIF).
The custom code can process the service response as appropriate—log the
results, display the data to the user in a separate Swing dialog (if custom-
coded and the custom function is client-side), allow users to copy and paste
the results into a data entry field, execute real-time PUT statements of the
data back into the correct business objects, and so on.
Custom buttons are displayed to the right of the top panel of the Merge
Manager, in the same location as the regular Merge Manager buttons. The
following example shows a button called fx.
Custom buttons are displayed in the top part of the top panel of the Hierarchy
Manager screen, in the same location as other Hierarchy Manager buttons.
The following example shows a button called fx.
- 731 -
2. Write and package the business logic that the custom button will execute,
as described in "Writing a Custom Function" on page 732.
3. Deploy the package so that it appears in the applicable tool(s) in the Hub
Console, as described in "Deploying Custom Buttons" on page 735.
Once an external service button is visible in the Hub Console, users can click
the button to invoke the service.
com.siperian.mrm.customfunctions.api.CustomFunction
To learn more about this interface, see the Javadoc that accompanies your
Informatica MDM Hub distribution.
This section provides the Java code for two example custom functions that
implement the com.siperian.mrm.customfunctions.api.CustomFunction
interface. The code simply prints (on standard error) information to the server
log or the Hub Console log.
The name of the client function class for the following sample code is
com.siperian.mrm.customfunctions.test.TestFunction.
//=====================================================================
//project: Informatica Master Reference Manager, Hierarchy Manager
//---------------------------------------------------------------------
//copyright: Informatica Corp. (c) 2008-2010. All rights reserved.
//=====================================================================
- 732 -
package com.siperian.mrm.customfunctions.test;
import java.awt.Frame;
import java.util.Properties;
import javax.swing.Icon;
import com.siperian.mrm.customfunctions.api.CustomFunction;
The name of the server function class for the following code is
com.siperian.mrm.customfunctions.test.TestFunctionClient.
//=====================================================================
//project: Informatica Master Reference Manager, Hierarchy Manager
//---------------------------------------------------------------------
//copyright: Informatica Corp. (c) 2008-2010. All rights reserved.
//=====================================================================
package com.siperian.mrm.customfunctions.test;
- 733 -
import java.awt.Frame;
import java.util.Properties;
import javax.swing.Icon;
import com.siperian.mrm.customfunctions.api.CustomFunction;
/**
* This is a sample custom function that is executed on the Server.
* To deploy this function, put it in a jar file and upload the jar file
* to the DB using DeployCustomFunction.
*/
public class TestFunction implements CustomFunction {
public String getActionText() {
return "Test Server";
}
public Icon getGuiIcon() {
return null;
}
- 734 -
Method Description
getActionText Specify the text for the button label. Uses the default visual
appearance for custom buttons.
getGuiIcon Specify the icon graphic in any Swing-compatible graphic
format (such as JPG, PNG, or GIF). This image file can be
bundled with the JAR file for this custom function.
Reply to the prompts based on the configured settings for your Informatica
MDM Hub implementation. For example:
Database Type:oracle
Host:localhost
Port(1521):
Service:orcl
Username:ds_ui1
Password:!!cmx!!
(L)ist, (A)dd, (U)pdate, (C)hange Type, (S)et Properties, (D)elete or
(Q)uit:l
No custom actions
(L)ist, (A)dd, (U)pdate Jar, (C)hange Type, (S)et Properties, (D)elete
or (Q)uit:q
- 735 -
4. When prompted, specify database connection information: database host,
port, service, login username, and password.
5. The DeployCustomFunction tool displays a menu of the following options.
Label Description
(L)ist Displays a list of currently-defined custom buttons.
(A)dd Adds a new custom button. The DeployCustomFunction tool
prompts you to specify:
• the JAR file for your custom button
• the name of the custom function class that implements the
com.siperian.mrm.customfunctions.api.CustomFunction
interface
• the type of the custom button: m—Merge Manager, d—Data
Manager, h—Hierarchy Manager (you can specify one or
two letters)
(U)pdate Updates the JAR file for an existing custom button. The
DeployCustomFunction tool prompts you to specify:
• the rowID of the custom button to update
• the JAR file for your custom button
• the name of the custom function class that implements the
com.siperian.mrm.customfunctions.api.CustomFunction
interface
• the type of the custom button: m—Merge Manager, h—
Hierarchy Manager (you can specify one or two letters)
(C)hange Changes the type of an existing custom button. The
Type DeployCustomFunction tool prompts you to specify:
• the rowID of the custom button to update
• the type of the custom button: m—Merge Manager, and /or
h—Hierarchy Manager (you can specify one or two letters)
(S)et Specify a properties file, which defines name/value pairs that
Properties
the custom function requires at execution time (name=value).
The DeployCustomFunction tool prompts you to specify the
properties file to use.
(D)elete Deletes an existing custom button. The DeployCustomFunction
tool prompts you to specify the rowID of the custom button to
delete.
(Q)uit Exits the DeployCustomFunction tool.
6. When you have finished choosing your actions, choose (Q)uit.
7. Refresh the browser window to display the custom button you just added.
8. Test your custom button to ensure that it works properly.
- 736 -
Appendix F: Configuring Access to
Hub Console Tools
Appendix Contents
• "About User Access to Hub Console Tools" on page 737
• "Starting the Tool Access Tool" on page 737
• "Granting User Access to Tools and Processes" on page 738
• "Revoking User Access to Tools and Processes" on page 739
You use the Tool Access tool in the Configuration workbench to configure
access to Hub Console tools. To use the Tool Access tool, you must be
connected to the master database.
Note: The Tool Access tool applies only to Informatica MDM Hub users who
are not configured as administrators (users who do not have the
Administrator check box selected in the Users tool, as described in "Editing
User Accounts" on page 649).
- 737 -
In the above example, the cmx_global user account exists only to store the
global password policy, which is described in "Managing the Global Password
Policy" on page 654.
- 738 -
The user will have access to these processes and tools for every ORS to which
they have access. You cannot give a user access to one tool for one ORS and
another tool for a different ORS.
Note: If you want to grant access to only some of the tools in a workbench,
then expand the associated workbench in the Accessible tools and
processes list, select the tool, and revoke access according to the
instructions in the next section, "Revoking User Access to Tools and
Processes" on page 739.
- 739 -
Appendix G: Row-level Locking
This appendix describes how to enable and use row-level locking to protect
data integrity when asynchronously running batch and API processes
(including SIF requests) or API/API processes.
Note: You can skip this section if batch jobs and API calls are not run
concurrently in your Informatica MDM Hub implementation.
Appendix Contents
• "About Row-level Locking" on page 740
• "Configuring Row-level Locking" on page 741
• "Locking Interactions Between SIF Requests and Batch Processes" on page
742
Default Behavior
Row-level locking is disabled by default. While disabled, API processes
(including SIF requests) and batch processes cannot be executed
asynchronously on the same base object at the same time. If you explicitly
enable row-level locking for an ORS (see "Enabling Row-level Locking on an
ORS" on page 741), then Informatica MDM Hub uses the Oracle row-level
locking mechanism to manage concurrent updates for tokenize, match, and
merge processes.
- 740 -
Types of Locks
Data management in Informatica MDM Hub involves the following types of
locks.
Name Definition
exclusive Prohibits all other jobs (API or batch processes) from processing on
lock the locked base object.
shared Prohibits only certain jobs from running. For example, a batch job
lock could issue a non-exclusive on a base object and, when
interoperability is enabled (on), this shared lock would prohibit
other batch jobs but would allow API jobs to attempt to process on
the base object.
row- Shared lock that also includes a SELECT FOR UPDATE to lock the
level affected base object rows.
lock
Note: Interoperabilty must be enabled if Batch jobs are being run together.
If, you have multiple parents that attempt access to the same child (or parent)
records when running different Batch jobs, one job will fail if it attempts to
lock records being processed by the other batch and it holds them longer than
the batch wait time. The maximum wait time is defined in C_REPOS_TABLE.
- 741 -
3. Edit the ORS properties according to the instructions in "Editing ORS
Properties" on page 64.
4. Select (check) the Batch API Lock Interoperability check box.
5. Save your changes.
Note: Once you enable row-level locking, the Tokenize on Put property cannot
be enabled.
The following table shows the interactions between the API and batch
processes when row-level locking is enabled.
Locking Interactions When API and Batch Interoperability is Enabled
Existing Batch – Batch – Shared Lock API Row-level Lock
Lock / Exclusive
New Lock
Incoming
Call
Batch – Immediately Immediately display An Wait for Batch_Lock_Wait_
Exclusive display an error message Seconds checking the lock
Lock error existence. Display an error
message message if the lock is not
cleared by the wait time.
Called for each table to be
locked.
Batch – Immediately Immediately display An Wait for Batch_Lock_Wait_
Shared display an error message Seconds to apply a row-level
Lock error lock using FOR UPDATE
message SELECT. If the table does not
manage the lock, display an
error message. Called for each
table to be locked.
- 742 -
Existing Batch – Batch – Shared Lock API Row-level Lock
Lock / Exclusive
New Lock
Incoming
Call
API - Immediately Wait for API_Lock_ Wait for API_Lock_Wait_
Row level display an Wait_Seconds to apply Seconds to apply a row-level
lock error a row-level lock using lock using FOR UPDATE
message FOR UPDATE SELECT. If SELECT. If the table does not
table does not manage manage the lock, display an
the lock, display An error message. Called for each
error message. table to be locked.
The following table shows the interactions between the API and batch
processes when API batch interoperability is disabled. In this scenario, batch
processes will issue an exclusive lock, while SIF requests will check for an
exclusive lock but will not issue any locks.
Locking Interactions When API and Batch Interoperability is Disabled
Existing Lock / New Batch API
Incoming Call
Batch – Exclusive Lock Immediately display an See "Default Behavior"
error message on page 740.
API Immediately display an See "Default Behavior"
error message on page 740.
- 743 -
Glossary
accept limit
This is a state associated with a base object or cross reference record. A base
object record is active if at least one of its cross reference records is active. A
cross reference record contributes to the consolidated base object only if it is
active.
Active records participate in Hub processes by default. These are the records
that are available to participate in any operation. If records are required to go
through an approval process, then these records have been through that
process and have been approved.
Activity Manager
Default source system. Used for manual trust overrides and data edits from
the Data Manager or Merge Manager tools. See "source system" on page 779.
- 744 -
administrator
Informatica MDM Hub user who has the primary responsibility for configuring
the Informatica MDM Hub system. Administrators access Informatica MDM
Hub through the Hub Console, and use Informatica MDM Hub tools to configure
the objects in the Hub Store, and create and modify Informatica MDM Hub
security.
auditable events
Informatica MDM Hub provides the ability to create an audit trail for certain
activities that are associated with the exchange of data between Informatica
MDM Hub and external systems. An audit trail can be captured whenever:
• an external application interacts with Informatica MDM Hub by invoking a
Services Integration Framework (SIF) request
• Informatica MDM Hub sends a message (using JMS) to a message queue
for the purpose of distributing data changes to other systems
• Activity Manager invokes an external application.
authentication
Process of verifying the identity of a user to ensure that they are who they
claim to be. In Informatica MDM Hub, users are authenticated based on their
supplied credentials—user name / password, security payload, or a
combination of both. Informatica MDM Hub provides an internal authentication
mechanism and also supports user authentication via third-party
authentication providers. See "credentials" on page 750, "security payload" on
page 778.
authorization
automerge
- 745 -
intervention. See "manual merge" on page 762, "merge-style base object" on
page 767.
base object
batch group
A collection of individual batch jobs (for example, Stage, Load, and Match
jobs) that can be executed with a single command. Each batch job in a group
can be executed sequentially or in parallel to other jobs. See also "batch job"
on page 746.
batch job
batch mode
Way of interacting with Informatica MDM Hub via batch jobs, which can be
executed in the Hub Console or using third-party management tools to
schedule and execute batch jobs (in the form of stored procedures) on the
database server. See also "real-time mode" on page 773, "batch job" on page
746, "batch group" on page 746, "stored procedure" on page 780.
A record that has been consolidated with the best cells of data from the source
records. Sometimes abbreviated as BVT.
• For merge-style base objects, the base object record is the BVT record,
and is built by consolidating the most-trustworthy cell values from the
corresponding source records.
BI vendor
bulk merge
- 746 -
BVT
cascade delete
When the Delete stored procedure deletes records in the parent object, it also
removes the affected records in the child base object. To enable a cascade
delete operation, set the CASCADE_DELETE_IND parameter to 1. The Delete job
checks each child base object table for related data that should be deleted
given the removal of the parent base object record.
If you do not set this parameter, Informatica MDM Hub generates an error
message if there are child base objects referencing the deleted base object
record; the Delete job fails, and Informatica MDM Hub performs a rollback
operation for the associated data.
cascade unmerge
When records in a parent object are unmerged, Informatica MDM Hub also
unmerges affected records in the child base object.
See also: "linear unmerge" on page 761, "tree unmerge" on page 782.
cell
change list
cleanse
cleanse engine
A cleanse engine is a third party product used to perform data cleansing with
the Informatica MDM Hub.
- 747 -
cleanse function
Code changes the incoming data during Stage jobs, converting each input
string to an output string. Typically, these functions are used to standardize
data and thereby optimize the match process. By combining multiple cleanse
functions, you can perform complex filtering and standardization. See also
"data cleansing" on page 751, "internal cleanse" on page 760.
cleanse list
A logical grouping of rules for replacing parts of an input string during the
cleanse process. See "cleanse function" on page 748, "data cleansing" on page
751.
The Cleanse Match Server is multi-threaded so that each instance can process
multiple requests concurrently. It can be deployed on a variety of application
servers.
The Cleanse Match Server interfaces with any of the supported cleanse
engines, such as the Trillium Director cleanse engine. The Cleanse Match
Server and the cleanse engine work to standardize the data. This
standardization works closely with the Informatica Consolidation Engine
(formerly referred to as the Merge Engine) to optimize the data for
consolidation.
column
In a table, a set of data values of a particular type, one for each row of the
table. See "system column" on page 780, "user-defined column" on page 783.
A change list that is the result of comparing the contents of two repositories
and generating the list of changes to make to the target repository.
Comparison change lists are used in Metadata Manager when promoting or
importing design objects. See also "change list" on page 747, "creation change
list" on page 750, "Metadata Manager" on page 768.
- 748 -
complete match tracking
The display of the complete or original match chain that caused two records to
be matched through intermediate records.
conditional mapping
A mapping between a column in a landing table and a staging table that uses a
SQL WHERE clause to conditionally select only those records in the landing
table that meet the filter condition. See "mapping" on page 762, "distinct
mapping" on page 753.
Configuration workbench
Includes tools for configuring a variety of Hub objects, including, the ORS,
users, security, message queues, and metadata validation.
consolidated record
consolidation process
Process of merging or linking duplicate records into a single record. The goal
in Informatica MDM Hub is to identify and eliminate all duplicate data and to
merge or link them together into a single, consolidated record while
maintaining full traceability.
consolidation indicator
- 749 -
Indicator State Name Description
Value
records. For more information, see the Informatica
MDM Hub Data Steward Guide.
content metadata
Data that describes the business data that has been processed by Informatica
MDM Hub. Content metadata is stored in support tables for a base object,
including cross-reference tables, history tables, and others. Content metadata
is used to help determine where the data in the base object came from, and
how the data changed over time.
control table
credentials
What a user supplies at login time to gain access to Informatica MDM Hub
resources. Credentials are used during the authorization process to determine
whether a user is who they claim to be. Login credentials might be a user
name and password, a security payload (such as a security token or some
other binary data), or a combination of user name/password and security
payload. See "authentication" on page 745, "security payload" on page 778.
cross-reference table
- 750 -
Customer Data Integration (CDI)
A discipline within "Master Data Management (MDM)" on page 763 that focuses
on customer master data and its related attributes. See "master data" on page
763.
database
Organized collection of data in the Hub Store. Informatica MDM Hub supports
two types of databases: a Master Database and an Operational Reference
Store (Operational Reference Store). See "Master Database" on page 763,
"Operational Reference Store (ORS)" on page 769, and "Hub Store" on page
759.
data cleansing
Data Manager
Use the Data Manager tool to search for records, view their cross-references,
unmerge records, unlink records, view history records, create new records,
edit records, and override trust settings. The Data Manager displays all
records that meet the search criteria you define.
datasource
- 751 -
server, the database name, the database user ID and password, and so on.
Informatica MDM Hub needs this information to communicate with an ORS.
data steward
Informatica MDM Hub user who has the primary responsibility for data quality.
Data stewards access Informatica MDM Hub through the Hub Console, and use
Informatica MDM Hub tools to configure the objects in the Hub Store.
Part of the Informatica MDM Hub UI used to review consolidated data as well
as matched data queued for exception handling by data analysts or stewards
who understand the data semantics and are guardians of data reliability in an
organization.
Includes tools for using the Data Manager, Merge Manager, and Hierarchy
Manager.
data type
decay curve
Visually shows the way that trust decays over time. Its shape is determined
by the configured decay type and decay period. See "decay period" on page
752, "decay type" on page 752.
decay period
The amount of time (days, weeks, months, quarters, and years) that it takes
for the trust level to decay from the maximum trust level to the minimum
trust level. See "decay curve" on page 752, "decay type" on page 752.
decay type
The way that the trust level decreases during the decay period. See "linear
decay" on page 761, "RISL decay" on page 775, "SIRL decay" on page 778,
"decay curve" on page 752, "decay period" on page 752.
- 752 -
deleted state (records)
Deleted records are records that are no longer desired to be part of the Hub’s
data. These records are not used in process (unless specifically requested).
Records can only be deleted explicitly and once deleted can be restored if
desired. When a record that is Pending is deleted, it is permanently deleted
and cannot be restored.
delta detection
During the stage process, Informatica MDM Hub only processes new or
changed records when this feature is enabled. Delta detection can be done
either by comparing entire records or via a date column.
design object
Parts of the metadata used to define the schema and other configuration
settings for an implementation. Design objects include instances of the
following types of Informatica MDM Hub objects: base objects and columns,
landing and staging tables, columns, indexes, relationships, mappings,
cleanse functions, queries and packages, trust settings, validation and match
rules, Security Access Manager definitions, Hierarchy Manager definitions,
and other settings. See "metadata" on page 768, "Metadata Manager" on page
768.
distinct mapping
A mapping between a column in a landing table and a staging table that selects
only the distinct records from the landing table. Using distinct mapping is
useful in situations in which you have a single landing table feeding multiple
staging tables and the landing table is denormalized (for example, it contains
both customer and address data). See "mapping" on page 762, "conditional
mapping" on page 749.
A source system that provides data that gets inserted into the base object
without being consolidated. See "source system" on page 779.
distribution
- 753 -
downgrade
Operation that occurs when inserting or updating data using the load process
or using cleansePut & Put APIs when a validation rule reduces the trust for a
record by a percentage.
duplicate
One or more records in which the data in certain columns (such as name,
address, or organization data) is identical or nearly identical. Match rules
executed during the match process determine whether two records are
sufficiently similar to be considered duplicates for consolidation purposes.
entity
entity type
In Hierarchy Manager, entity types define the kinds of objects that can be
related using Hierarchy Manager. Examples are individual, organization,
product, and household. All entities with the same entity type are stored in the
same entity base object. In the HM Configuration tool, entity types are
displayed in the navigation tree under the Entity Object with which the Type is
associated. See "entity" on page 754.
exact match
A match / search strategy that matches only records that are identical. If you
specify an exact match, you can define only exact match columns for this base
object (exact-match base objects cannot have fuzzy match columns). A base
object that uses the exact match / search strategy is called an exact-match
base object. See also "match / search strategy" on page 766, "fuzzy match" on
page 756.
- 754 -
exclusive lock
In the Hub Console, a lock that is required in order to make exclusive changes
to the underlying schema. An exclusive lock prevents all other Hub Console
users from making changes to the target database at the same time. An
exclusive lock must be released by the user with the exclusive lock; it cannot
be cleared by another user. See "write lock" on page 785.
execution path
The sequence in which batch jobs are executed when the entire batch group is
executed in the Informatica MDM Hub. The execution path begins with the
Start node and ends with the End node. The Batch Group tool does not validate
the execution sequence for you—it is up to you to ensure that the execution
sequence is correct.
export process
Informatica MDM Hub user who access Informatica MDM Hub data indirectly
via third-party applications.
external cleanse
The process of cleansing data prior to populating the landing tables. External
cleansing is typically performed outside of Informatica MDM Hub using an
extract-transform-load (ETL) tool or some other data cleansing utility. See
also "data cleansing" on page 751, "extract-transform-load (ETL) tool" on page
756, "internal cleanse" on page 760.
external match
Process that allows you to match new data (stored in a separate input table)
with existing data in a fuzzy-match base object, test for matches, and inspect
the results—all without actually changing data in the base object in any way,
or changing the match table associated with the base object.
- 755 -
extract-transform-load (ETL) tool
A software tool (external to Informatica MDM Hub) that extracts data from a
source system, transforms the data (using rules, lookup tables, and other
functionality) to convert it to the desired state, and then loads (writes) the
data to a target database. For Informatica MDM Hub implementations, ETL
tools are used to extract data from source systems and populate the landing
tables. See also "data cleansing" on page 751, "external cleanse" on page 755.
foreign key
function
fuzzy match
A match / search strategy that uses probabilistic matching, which takes into
account spelling variations, possible misspellings, and other differences that
can make matching records non-identical. If selected, Informatica MDM Hub
adds a special column (Fuzzy Match Key) to the base object. A base object
that uses the fuzzy match / search strategy is called a fuzzy-match base
object. Using fuzzy match requires a selected population. See "fuzzy match
key" on page 756, "match / search strategy" on page 766, "exact match" on
page 754, and "population" on page 770.
Special column in the base object that the Schema Manager adds if a match
column uses the fuzzy match / search strategy. This column is the primary
field used during searching and matching to generate match candidates for
this base object. All fuzzy base objects have one and only one Fuzzy Match
Key. See "fuzzy match" on page 756, "match key" on page 764, "match key
table" on page 764.
A column that contains common identifiers (key values) that allow you to
uniquely and globally identify a record based on your business needs.
Examples include:
- 756 -
• identifiers defined by applications external to Informatica MDM Hub, such
as ERP or CRM systems.
• Identifiers defined by external organizations, such as industry-specific
codes (AMA numbers, DEA numbers. and so on), or government-issued
identifiers (social security number, tax ID number, driver’s license
number, and so on).
hard delete
Hierarchies Tool
Hierarchy Manager
Part of the Informatica MDM Hub UI used to set up the structures required to
view and manipulate data relationships. Informatica Hierarchy Manager
(Hierarchy Manager or HM) builds on Informatica Master Reference Manager
(MRM) and the repository managed by Informatica MDM Hub for reference and
relationship data. Hierarchy Manager gives you visibility into how
relationships correlate between systems, enabling you to discover
opportunities for more effective customer service, to maximize profits, or to
enact compliance with established standards.
The Hierarchy Manager tool is accessible via the Data Steward workbench.
hierarchy
- 757 -
hierarchy type
history table
HM package
hotspot
Hub Console
Informatica MDM Hub user interface that comprises a set of tools for
administrators and data stewards. Each tool allows users to perform a specific
action, or a set of related actions, such as building the data model, running
batch jobs, configuring the data flow, running batch jobs, configuring external
application access to Informatica MDM Hub resources, and other system
configuration and operation tasks.
hub object
A generic term for various types of objects defined in the Hub that contain
information about your business entities. Some examples include: base
objects, cross reference tables, and any object in the hub that you can
associate with reporting metrics.
Hub Server
A run-time component in the middle tier (application server) used for core and
common services, including access, security, and session management.
- 758 -
Hub Store
immutable source
A data source that always provides the best, final version of the truth for a
base object. Records from an immutable source will be accepted as unique
and, once a record from that source has been fully consolidated, it will not be
changed—even in the event of a merge. Immutable sources are also distinct
systems. For all source records from an immutable source system, the
consolidation indicator for Load and PUT is always 1 (consolidated record).
implementer
Informatica MDM Hub user who has the primary responsibility for designing,
developing, testing, and deploying Informatica MDM Hub according to the
requirements of an organization. Tasks include (but are not limited to)
creating design objects, building the schema, defining match rules,
performance tuning, and other activities.
import process
incremental load
Any load process that occurs after a base object has undergone its initial data
load. Called incremental loading because only new or updated data is loaded
into the base object. Duplicate data is ignored. See "initial data load" on page
760.
indirect match
- 759 -
initial data load
The very first time that you data is loaded into an empty base object. During
the initial data load, all records in the staging table are inserted into the base
object as new records.
internal cleanse
The process of cleansing data during the stage process, when data is copied
from landing tables to the appropriate staging tables. Internal cleansing
occurs inside Informatica MDM Hub using configured cleanse functions that
are executed by the Cleanse Match Server in conjunction with a supported
cleanse engine. See also "data cleansing" on page 751, "cleanse engine" on
page 747, "external cleanse" on page 755.
In the Batch Viewer and Batch Group tools, a log that shows job completion
status with any associated messages, such as success, failure, or warning.
A Informatica MDM Hub batch job that matches records from two or more
sources when these sources use the same primary key. Key Match jobs
compare new records to each other and to existing records, and then identify
potential matches based on the comparison of source record keys as defined
by the primary key match rules. See "primary key match rule" on page 771,
"match process" on page 765.
key type
key width
During match, determines how fast searches are during match, the number of
possible match candidates returned, and how much disk space the keys
- 760 -
consume. Key width options are Standard, Extended, Limited, and Preferred.
Key widths apply to fuzzy match objects only. See "match process" on page
765.
land process
landing table
A table where a source system puts data that will be processed by Informatica
MDM Hub.
lineage
linear decay
The trust level decreases in a straight line from the maximum trust to the
minimum trust. See "decay type" on page 752, "trust" on page 782.
linear unmerge
A base object record is unmerged and taken out of the existing merge tree
structure. Only the unmerged base object record itself will come out the
merge tree structure, and all base object records below it in the merge tree
will stay in the original merge tree.
See also: "cascade unmerge" on page 747, "tree unmerge" on page 782.
load insert
When records are inserted into the target base object. During the load
process, if a record in the staging table does not already exist in the target
table, then Informatica MDM Hub inserts the record into the target table. See
"load process" on page 761, "load update" on page 762.
load process
Process of loading data from a staging table into the corresponding base
object in the Hub Store. If the new data overlaps with existing data in the Hub
Store, Informatica MDM Hub uses trust settings and validation rules to
- 761 -
determine which value is more reliable. See "trust" on page 782, "validation
rule" on page 785, "load insert" on page 761, "load update" on page 762.
load update
When records are inserted into the target base object. During the load
process, if a record in the staging table does not already exist in the target
table, then Informatica MDM Hub inserts the record into the target table. See
"load process" on page 761, "load insert" on page 761.
lock
lookup
Process of retrieving a data value from a parent table during Load jobs. In
Informatica MDM Hub, when configuring a staging table associated with a base
object, if a foreign key column in the staging table (as the child table) is
related to the primary key in a parent table, you can configure a lookup to
retrieve data from that parent table.
manual merge
manual unmerge
mapping
Defines a set of transformations that are applied to source data. Mappings are
used during the stage process (or using the SiperianClient CleansePut API
request) to transfer data from a landing table to a staging table. A mapping
identifies the source column in the landing table and the target column to
populate in the staging table, along with any intermediate cleanse functions
used to clean the data. See "conditional mapping" on page 749, "distinct
mapping" on page 753.
- 762 -
master data
The controlled process by which the master data is created and maintained as
the system of record for the enterprise. MDM is implemented in order to
ensure that the master data is validated as correct, consistent, and complete,
and—optionally—circulated in context for consumption by internal or external
business processes, applications, or users. See "master data" on page 763,
"Customer Data Integration (CDI)" on page 751.
Master Database
master record
Single record in the base object that represents the “best version of the truth”
for a given entity (such as a specific organization or person). The master
record represents the fully-consolidated data for the entity.
- 763 -
match
match candidate
For fuzzy-match base objects only, any record in the base object that is a
possible match.
match column
A column that is used in a match rule for comparison purposes. Each match
column is based on one or more columns from the base object. See "match
process" on page 765.
Match rule that is used to match records based on the values in columns you
have defined as match columns, such as last name, first name, address1, and
address2. See "primary key match rule" on page 771, "match process" on
page 765.
match key
Encoded strings that represent the data in the fuzzy match key column of the
base object. Match keys consist of fixed-length, compressed, and encoded
values built from a combination of the words and numbers in a name or
address such that relevant variations have the same match key value. Match
keys are one part of the match tokens that are generated during the tokenize
process, stored in the match key table, and then used during the match
process to identify candidates for matching. See "match token" on page 766,
"fuzzy match key" on page 756, "match key table" on page 764, "tokenize
process" on page 781, "match process" on page 765.
System table that stores the match tokens (match keys + unencoded, raw
data) that are generated during the tokenize process. This data is used during
the match process to identify candidates for matching, comparing the match
keys according to the match rules that have been defined to determine which
records are duplicates. See "match key" on page 764, "match token" on page
766, "tokenize process" on page 781, "match process" on page 765.
- 764 -
match list
match path
match process
match purpose
For fuzzy-match base objects, defines the primary goal behind a match rule.
For example, if you're trying to identify matches for people where address is
an important part of determining whether two records are for the same
person, then you would use the Match Purpose called Resident. Each match
purpose contains knowledge about how best to compare two records to
achieve the purpose of the match. Informatica MDM Hub uses the selected
match purpose as a basis for applying the match rules to determine matched
records. The behavior of the rules is dependent on the selected purpose. See
"match process" on page 765.
match rule
A logical collection of match rules that allow users to execute different sets of
rules at different stages in the match process. Match rule sets include a search
level that dictates the search strategy, any number of automatic and manual
match rules, and optionally, a filter that allows you to selectively include or
- 765 -
exclude records during the match process Match rules sets are used to
execute to match column rules but not primary key match rules. See "match
process" on page 765.
match subtype
Used with base objects that containing different types of data, such as an
Organization base object containing customer, vendor, and partner records.
Using match subtyping, you can apply match rules to specific types of data
within the same base object. For each match rule, you specify an exact match
column that will serve as the “subtyping” column to filter out the records that
you want to ignore for that match rule. See "match process" on page 765.
match table
Type of system table, associated with a base object, that supports the match
process. During the execution of a Match job for a base object, Informatica
MDM Hub populates its associated match table with the ROWID_OBJECT values
for each pair of matched records, as well as the identifier for the match rule
that resulted in the match, and an automerge indicator. See "match process"
on page 765.
match token
Strings that represent both encoded (match key) and unencoded (raw) values
in the match columns of the base object. Match tokens are generated during
the tokenize process, stored in the match key table, and then used during the
match process to identify candidates for matching. See "match key" on page
764, "match key table" on page 764, "match process" on page 765, "tokenize
process" on page 781.
match type
Each match column has a match type that determines how the match column
will be tokenized in preparation for the match comparison. See "match
process" on page 765.
Specifies the reliability of the match versus the performance you require:
fuzzy or exact. An exact match / search strategy is faster, but an exact match
will miss some matches if the data is imperfect. See "fuzzy match" on page
756, "exact match" on page 754., "match process" on page 765.
- 766 -
maximum trust
The trust level that a data value will have if it has just been changed. For
example, if source system A changes a phone number field from 555-1234 to
555-4321, the new value will be given system A’s maximum trust level for the
phone number field. By setting the maximum trust level relatively high, you
can ensure that changes in the source systems will usually be applied to the
base object.
merge process
Process of combining two or more records of a base object table because they
have the same value (or very similar values) in the specified match columns.
See "consolidation process" on page 749, "automerge" on page 745, "manual
merge" on page 762, "manual unmerge" on page 762.
Type of base object that is used with Informatica MDM Hub’s match and merge
capabilities. See "merge process" on page 767.
Merge Manager
Tool used to review and take action on the records that are queued for manual
merging.
message
message queue
A mechanism for transmitting data from one process to another (for example,
from Informatica MDM Hub to an external application).
A mechanism for identifying base object events and transferring the effected
records to the internal system for update. Message queue rules are supported
for updates, merges, and records accepted as unique.
- 767 -
message queue server
message trigger
A rule that gets fired when which a particular action occurs within Informatica
MDM Hub. When an action occurs for which a rule is defined, a JMS message is
placed in the outbound message queue. A message trigger identifies the
conditions which cause the message to be generated (what action on which
object) and the queue on which messages are placed.
metadata
Data that is used to describe other data. In Informatica MDM Hub, metadata is
used to describe the schema (data model) that is used in your Informatica
MDM Hub implementation, along with related configuration settings. See also
"Metadata Manager" on page 768, "design object" on page 753, "schema" on
page 776.
Metadata Manager
The Metadata Manager tool in the Hub Console is used to validate metadata for
a repository, promote design objects from one repository to another, import
design objects into a repository, and export a repository to a change list. See
also "metadata" on page 768, "design object" on page 753, "validation
process" on page 784, "import process" on page 759, "promotion process" on
page 772, "export process" on page 755, "change list" on page 747.
metadata validation
minimum trust
The trust level that a data value will have when it is “old” (after the decay
period has elapsed). This value must be less than or equal to the maximum
trust. If the maximum and minimum trust are equal, the decay curve is a flat
line and the decay period and decay type have no effect. See also "decay
period" on page 752.
- 768 -
Model workbench
Part of the Informatica MDM Hub UI used to configure the solution during
deployment by the implementers, and for on-going configuration by data
architects of the various types of metadata and rules in response to changing
business needs.
Includes tools for creating query groups, defining packages and other schema
objects, and viewing the current schema.
A cross-reference (XREF) record that does not contribute to the BVT (best
version of the truth) of the base object record. As a consequence, the values
in the cross-reference record will never show up in the base object record.
Note that this is for state-enabled records only.
non-equal matching
null value
The absence of a value in a column of a record. Null is not the same as blank
or zero.
operation
Database that contains the rules for processing the master data, the rules for
managing the set of master data objects, along with the processing rules and
auxiliary logic used by the Informatica MDM Hub in defining the BVT. A
Informatica MDM Hub configuration can have one or more ORS databases.
The default name of an ORS is CMX_ORS. See also "Master Database" on page
763.
overmatching
For fuzzy-match base objects only, a match that results in too many matches,
including matches that are not relevant. When configuring match, the goal is
- 769 -
to find the optimal number of matches for your data. See "undermatching" on
page 783.
package
password policy
path
Pending records are records that have not yet been approved for general
usage in the Hub. These records can have most operations performed on
them, but operations have to specifically request Pending records. If records
are required to go through an approval process, then these records have not
yet been approved and are in the midst of an approval process.
Specific security check points that determine, at run time, the validity of a
user’s identity ("authentication" on page 745), along with that user’s access to
Informatica MDM Hub resources ("authorization" on page 745).
Specific security check points that enforce, at run time, security policies for
authentication and authorization requests.
population
Defines certain characteristics about data in the records that you are
matching. By default, Informatica MDM Hub comes with the US population, but
- 770 -
Informatica provides a standard population per country. Populations account
for the inevitable variations and errors that are likely to exist in name,
address, and other identification data; specify how Informatica MDM Hub
builds match tokens; and specify how search strategies and match purposes
operate on the population of data to be matched. Used only with the Fuzzy
match/search strategy.
primary key
Match rule that is used to match records from two systems that use the same
primary keys for records. See also "match column rule" on page 764.
private resource
A Informatica MDM Hub resource that is hidden from the Roles tool,
preventing its access via Services Integration Framework (SIF) operations.
When you add a new resource in Hub Console (such as a new base object), it
is designated a PRIVATE resource by default. See also "secure resource" on
page 777, "resource" on page 775.
privilege
- 771 -
profile
promotion process
provider
provider property
A name-value pair that a security provider might require in order to access for
the service(s) that they provide.
publish
query
A request to retrieve data from the Hub Store. Informatica MDM Hub allows
administrators to specify the criteria used to retrieve that data. Queries can
be configured to return selected columns, filter the result set with a WHERE
clause, use complex query syntax (such as GROUP BY, ORDER BY, and
HAVING clauses), and use aggregate functions (such as SUM, COUNT, and
AVG).
- 772 -
query group
raw table
real-time mode
reconciliation
For a given entity, Informatica MDM Hub obtains data from one or more
source systems, then reconciles “multiple versions of the truth” to arrive at
the master record—the best version of the truth—for that entity.
Reconciliation can involve cleansing the data beforehand to optimize the
process of matching and consolidating records for a base object. See
"distribution" on page 753.
record
referential integrity
regular expression
- 773 -
reject table
A table that contains records that Informatica MDM Hub could not insert into a
target table, such as:
• staging table (stage process) after performing the specified cleansing on a
record of the specified landing table
• Hub store table (load process)
A record could be rejected because the value of a cell is too long, or because
the record’s update date is later than the current date.
relationship
relationship type
repository
An Operational Reference Store (ORS). The ORS stores metadata about its
own schema and related property settings. In Metadata Manager, when
copying metadata between repositories, there is always a source repository
that contains the design object to copy, and the target repository that is
destination for the design object. See also "Metadata Manager" on page 768,
"validation process" on page 784, "import process" on page 759, "promotion
process" on page 772, "export process" on page 755, "change list" on page
747.
- 774 -
request
Informatica MDM Hub request (API) that allows external applications to access
specific Informatica MDM Hub functionality using the Services Integration
Framework (SIF), a request/response API model.
resource
Any Informatica MDM Hub object that is used in your Informatica MDM Hub
implementation. Certain resources can be configured as secure resources:
base objects, mappings, packages, remote packages, cleanse functions, HM
profiles, the audit table, and the users table. In addition, you can configure
secure resources that are accessible by SIF operations, including content
metadata, match rule sets, metadata, batch groups, the audit table, and the
users table. See "private resource" on page 771, "secure resource" on page
777, "resource group" on page 775.
resource group
Resource Kit
The Informatica MDM Hub Resource Kit is a set of utilities, examples, and
libraries that provide examples of Informatica MDM Hub functionality that can
be expanded on and implemented.
RISL decay
Rapid Initial Slow Later decay puts most of the decrease at the beginning of
the decay period. The trust level follows a concave parabolic curve. If a
source system has this decay type, a new value from the system will probably
be trusted but this value will soon become much more likely to be overridden.
role
row
- 775 -
rule
rule set
Ability to exclude records from being processed by a match rule set. For
example, if you had an Organization base object that contained multiple types
of organizations (customers, vendors, prospects, partners, and so on), you
could define a match rule set that selectively processed only vendors. See
"match process" on page 765.
schema
Schema Manager
The Schema Viewer tool is a design-time component in the Hub Console used
to visualize the schema configured for your Informatica MDM Hub
implementation. The Schema Viewer is particularly helpful for visualizing a
complex schema.
search levels
Defines how stringently Informatica MDM Hub searches for matches: narrow,
typical, exhaustive, or extreme. The goal is to find the optimal number of
matches for your data—not too few (undermatching), which misses significant
matches, or too many (overmatching), which generates too many matches,
including insignificant ones. See "overmatching" on page 769,
"undermatching" on page 783.
- 776 -
secure resource
A protected Informatica MDM Hub resource that is exposed to the Roles tool,
allowing the resource to be added to roles with specific privileges. When a
user account is assigned to a specific role, then that user account is authorized
to access the secure resources via SIF according to the privileges associated
with that role. In order for external applications to access a Informatica MDM
Hub resource using SIF operations, that resource must be configured as
SECURE. Because all Informatica MDM Hub resources are PRIVATE by default,
you must explicitly make a resource SECURE after the resource has been
added. See also "private resource" on page 771, "resource" on page 775.
Status Description
Setting
SECURE Exposes this Informatica MDM Hub resource to the Roles tool,
allowing the resource to be added to roles with specific privileges.
When a user account is assigned to a specific role, then that user
account is authorized to access the secure resources using SIF
requests according to the privileges associated with that role.
PRIVATE Hides this Informatica MDM Hub resource from the Roles tool.
Default. Prevents its access via Services Integration Framework
(SIF) operations. When you add a new resource in Hub Console
(such as a new base object), it is designated a PRIVATE resource by
default.
security
- 777 -
security provider
security payload
Raw binary data supplied to a Informatica MDM Hub operation request that
can contain supplemental data required for further authentication and/or
authorization.
segment matching
The part of Informatica MDM Hub that interfaces with client programs.
Logically, it serves as a middle tier in the client/server model. It enables you
to implement the request/response interactions using any of the following
architectural variations:
• Loosely coupled Web services using the SOAP protocol.
• Tightly coupled Java remote procedure calls based on Enterprise
JavaBeans (EJBs) or XML.
• Asynchronous Java Message Service (JMS)-based messages.
• XML documents going back and forth via Hypertext Transfer Protocol
(HTTP).
SIRL decay
Slow Initial Rapid Later decay puts most of the decrease at the end of the
decay period. The trust level follows a convex parabolic curve. If a source
system has this decay type, it will be relatively unlikely for any other system
to override the value that it sets until the value is near the end of its decay
period.
soft delete
- 778 -
source record
A raw record from a source system. See also "record" on page 773, "source
system" on page 779.
source system
An external system that provides data to Informatica MDM Hub. See "distinct
source system" on page 753, "source record" on page 779.
stage process
Process of reading the data from the landing table, performing any configured
cleansing, and moving the cleansed data into the corresponding staging table.
If you enable delta detection, Informatica MDM Hub only processes new or
changed records. See "staging table" on page 779, "landing table" on page
761.
staging table
A table where cleansed data is temporarily stored before being loaded into
base objects via load jobs. See "stage process" on page 779, "load process" on
page 761.
state management
The process for managing the system state of base object and cross-reference
records to affect the processing logic throughout the MRM data flow. You can
assign a system state to base object and cross-reference records at various
stages of the data flow using the Hub tools that work with records. In addition,
you can use the various Hub tools for managing your schema to enable state
management for a base object, or to set user permissions for controlling who
can change the state of a record.
Rules that determine whether and when a record can change from one state to
another. State transition rules differ for base object and cross-reference
records.
- 779 -
stored procedure
stripping
strip table
When evaluating cells to merge from two records, Informatica MDM Hub
determines which cell data should survive and which one should be discarded.
The surviving cell data (or winning cell) is considered to represent the better
version of the truth between the two cells. Ultimately, a single, consolidated
record contains the best surviving cell data and represents the best version of
the truth.
survivorship
system column
- 780 -
system state
Describes how base object records are supported by Informatica MDM Hub.
The following states are supported: ACTIVE, PENDING, and DELETED. See
"state management" on page 779.
Systems and Trust tool is a design-time tool used to name the source systems
that can provide data for consolidation in Informatica MDM Hub. You use this
tool to define the trust settings associated with each source system for each
trust-enabled column in a base object.
table
target database
tokenize process
token table
- 781 -
traceability
The maintenance of data so that you can determine which systems—and which
records from those systems—contributed to consolidated records.
transactional data
transitive match
During the Build Match Group (BMG) process, a match that is made indirectly
due to the behavior of other matches. For example, if record 1 matches to
record 2, record 2 matches to record 3, and record 3 matches to record 4,
after the BMG process removes redundant matches, it might produce results
in which records 2, 3, and 4 match to record 1. In this example, there was no
explicit rule that matched record 4 to record 1. Instead, the match was made
indirectly.
tree unmerge
See also: "cascade unmerge" on page 747, "linear unmerge" on page 761.
trust
Mechanism for measuring the confidence factor associated with each cell
based on its source system, change history, and other business rules. Trust
takes into account the age of data, how much its reliability has decayed over
time, and the validity of the data.
- 782 -
trust level
For a source system that provides records to Informatica MDM Hub, a number
between 0 and 100 that assigns a level of confidence and reliability to that
source system, relative to other source systems. The trust level has meaning
only when compared with the trust level of another source system.
trust score
undermatching
For fuzzy-match base objects only, a match that results in too few matches,
which misses relevant matches. When configuring match, the goal is to find
the optimal number of matches for your data. See "overmatching" on page
769.
unmerge
user
user-defined column
Any column in a table that is not a system column. User-defined columns are
added in the Schema Manager and usually contain business data. See
"column" on page 748, "system column" on page 780.
- 783 -
user exit
user group
user object
Utilities workbench
Includes tools for auditing application event, configuring and running batch
groups, and generating the SIF APIs.
validation process
- 784 -
validation rule
Rule that tells Informatica MDM Hub the condition under which a data value is
not valid. When data meets the criteria specified by the validation rule, the
trust value for that data is downgraded by the percentage specified in the
validation rule. If the Reserve Minimum Trust flag is set for the column, then
the trust cannot be downgraded below the column’s minimum trust.
workbench
write lock
In the Hub Console, a lock that is required in order to make changes to the
underlying schema. All non-data steward tools (except the ORS security tools)
are in read-only mode unless you acquire a write lock. Write locks allow
multiple, concurrent users to make changes to the schema. See "exclusive
lock" on page 755.
- 785 -
Index: Accept Non-Matched Records As Unique jobs – base objects
events 684
- 786 -
Index: batch groups – batch jobs
described 74 ,
568
editing 95
asynchronous execution 563
exact match base objects 247
Auto Match and Merge jobs 532,
fuzzy match base objects 248
570
history table 89
Autolink jobs 532, 570
impact analysis 100
automatically-created batch jobs 499
load inserts 233
Automerge jobs 534, 571
load updates 234
BVT Snapshot jobs 535
overview of 83
C_REPOS_JOB_CONTROL table 565
record survivorship, state management162
C_REPOS_JOB_METRIC table 565
relationship base objects 374
C_REPOS_JOB_METRIC_TYPE table565
reserved suffixes 77
C_REPOS_JOB_STATUS_TYPEC table565
reverting from relationship base objects200
C_REPOS_TABLE_OBJECT_V table563
style 93
clearing history 512
system columns 84
command buttons 504
batch groups
configurable options 504
about batch groups 598
configuring 496
adding 514
design considerations 498
cmxbg.execute_batchgroup stored procedure599
executing 505
cmxbg.get_batchgroup_status stored procedure602
executing, about 559
cmxbg.reset_batchgroup stored procedure601
execution scripts 560
deleting 515
External Match jobs 535, 572
editing 515
foreign key relationships and 498
executing 522
Generate Match Token jobs 573
executing with stored procedures 598
Generate Match Tokens jobs 540
levels, configuring 516
Hub Delete jobs 541
stored procedures for 599
job execution logs 506
batch jobs
job execution status 506
about batch jobs 496
Key Match jobs 541, 579
Accept Non-Matched Records As Unique 532
Load jobs 542, 580
- 787 -
Index: Batch Viewer tool – cleanse functions
Manual Unlink jobs 546 best version of the truth (BVT) 82,
261
Manual Unmerge jobs 546
build match groups (BMGs) 249
Match Analyze jobs 550, 588
build_war macro 614
Match for Duplicate Data jobs 552,
589 BVT Snapshot jobs 535
- 788 -
Index: Cleanse Functions tool – cleansing data
- 789 -
Index: clearing history of batch jobs – custom buttons
- 790 -
Index: custom functions – Databases tool
- 791 -
Index: datasources – exact matches
datasources E
creating 71 entities
- 792 -
Index: exclusive locks – group execution logs
about External Match jobs 535 fuzzy match / search strategy 409
extreme search level 402 Generate Match Tokens jobs 540, 573
- 793 -
Index: hierarchies – Hub Store
H deleting 209
history toolbar 45
- 794 -
Index: HUB_STATE_IND column – landing tables
index L
- 795 -
Index: LAST_ROWID_SYSTEM column – mapping
expiration 35
- 796 -
Index: mappings – match columns
match strategy
match only previous rowid objects setting371exact match base objects 396
- 797 -
Index: Match for Duplicate Data jobs – match rules
Migrate Link Style to Merge Style jobs552 editing the name 406
- 798 -
Index: match subtype – messages
- 799 -
Index: metadata – Operational Reference Stores (ORS)
merge update message 471 Migrate Link Style to Merge Style jobs552
- 800 -
Index: operational reference stores (ORSs) – PKEY_SRC_OBJECT column
- 801 -
Index: populations – publish process
overview 260
- 802 -
Index: purposes, match – Revalidate jobs
- 803 -
Index: roles – security
roles starting 81
- 804 -
Index: Security Access Manager (SAM) – staging tables
list of provider files 666 Systems and Trust tool, starting 265
- 805 -
Index: standard key widths – table columns
- 806 -
Index: tables – trust
selecting 31 training 14
- 807 -
Index: typical search level – User Object Registry
U types 708
unmerge child when parent unmerges446 custom button functions, viewing 683
- 808 -
Index: user objects – write lock
- 809 -
Index: write locks – XSD file
write locks
exclusive locks 34
non-exclusive locks 34
XSD file
downloading 618
- 810 -