SAP HANA

SAP HANA

SAP HANA

4
5

SAP HANA

10

SQLScript

13

16 SAP HANA
17

SAP
SAP HANA

SAP
SAP HANA
SAP HANA(OLTP)
(OLAP)

SAP HANA

SAP
(side-by-side)

SAP HANA(OLTP)
(OLAP)

I/O

(OLAP)

Plattener Zeier

OLTP

1
(ERP)

BI

BI

1:

Corporate business intelligence (BI)

Enterprise data warehouse

Database

Local BI

Data mart

SAP ERP

SAP ERP

(or SAP CRM, SAP SRM,

SAP SCM)

(or SAP CRM, SAP SRM,

SAP SCM)

BI

Data mart

Database

Database

ETL*
Database

Non-SAP
applications

BI

Data mart

ETL
Database

Database

Database

*ETL = extract, transform, load

SAP HANA

CPU

2012 RAM 2 TB

2012 80
128 CPU

2: :

cPu
Core
CPU cache
Performance bottleneck
today: CPU waiting for data
to be loaded from memory

CPU 2

into cache

Main memory

Nehalem
CPU

Performance bottleneck in the

past: Disk I/O

Disk

L1 cache

64 KB

~4 cycles [2 ns]

L2 cache

256 KB

~10 cycles [5 ns]

L3 cache (shared)

8 MB

3540+ cycles [20 ns]

Main memory

GBs up to terabytes

100400 cycles

Solid state memory

GBs up to terabytes

5,000 cycles

Disk

Up to petabytes

1,000,000 cycles

OLTP
3SAPERP
2
CPU
CPU
CPU

CPU

3: SAP ERP2

Presentation
Native client

Mobile client

Web client

Business logic
Dispatcher and request queue management

Shared memory and caches

Work process 1

Work process n

Application server

Persistence
Database management system

SAP HANA

SAP HANA

SAP HANA
DBMSSQL (ACID

SAP HANA

SAP HANA4
SAP HANA SAP

[) ] SAP
HANASQL92 Open SQLSAP
SAP HANASQLSAP HANA
SQL
SAP HANA

SQLSAP HANAMDX
Microsoft Excel
BICSSAPBusiness Objects

SAP HANA

4: SAP HANA

SQL

SQLScript

MDX*

Other

App extensions
Business function library

Search

Predictive analysis library

Parallel calculation engine

Columnar

5SAP
16.832
1.420

relational stores
Row based

SAP HANASQL

SAP HANA

Objects graph store

SQL Script_
LSAP HANA
SAP

Managed appliance
*MDX = multidimensional expression


SAPSAP HANA

SQL

CPU SAP HANA

CPU

SAP HANACPU

SAP HANA

SAP HANA
SAP
SAP HANA

64SAP HANA on 4S Nehalem-EX (2.26 GHz) TPC-H (120,000,000,000)

16,822
Processing time
(milliseconds)
10,000

8,598

20% improvement due

to hyper threading with
64 logical cores
4,410
2,484
1,339

1,116

1,000
1

16

32

64

SAP HANA for Next-Generation Business Applications and real-time Analytics

6SQLMDX
SAP HANA

6: ()

SQLScript

MDX* query

Planning model

Other language/model

SQLScript compiler

MDX compiler

Planning engine

Other compiler

Standard SQL statement

calculation model (data flow graph)

calculation engine
Model optimizer (rule based)
R*

SQL processor

Intermediate results

Model executor
R
Calculation engine
operators

R
R

Logical
execution plan

Database optimizer
Statistics

Physical
execution plan

Script execution runtime

Database executor
R
Row store
*MDX = multidimensional expression; *R = request

Intermediate results
R

Column store

Execute user-defined
function

SQLScript

7:

document

8 SQLScript SAP HANA

SAP HANA SQLScript
SQL
SQLScript LC++
SAP C ++
BFLBFL

Relational
operation

Relational
operation

Relational
operation
Procedural
operation

SQLScript
SQLScript SQL
L
L SQL
SQL

SAP HANA

Relational
operation

Relational
operation

Relational
operation
Parallel 1

Parallel 2

Parallel 3

8: SAP HANA

SQL SQL

SAP HANA

Application server

SQLScript

c++
Business function library (BFL)

SAP HANA

11


SQL

SQL SQL

SAP HANA

SQL

SQLScript L

SAP HANA

9
1.
2.
3. 1

9: SAP HANA
table
Aggregation
thread 1

Aggregation
thread 2

Aggregation
thread 1
local hash table 1

Cache-sized hash tables

local hash table 1

Buffer

Merger
thread 1

Part hash table 1

Part hash table 2

Merger
thread 2

SAP HANA

SAP HANA

10

SAP HANA

selectupdate

SAP HANA
ABAP

10:

table
country

Product

Sales

US

Alpha

3,000

US

Beta

1,250

JP

Alpha

700

UK

Alpha

450

Row 1

Row 2

US

US

3,000

JP

US

UK
Product

Alpha

1,250

Beta

JP

Alpha

Alpha

Alpha

700

Sales

3,000

UK

1,250

Alpha

700

450

450

Row 4

US

Alpha

Beta

Row 3

Country

SAP HANA

13

SAP HANA

CPU

SAP

11
A B C

CPU

Temporal Tables

SAP HANA

SAP HANA

10

CPU
CPU

 11:

core 1

core 2

Processed by

Processed by

column A

column B

column c

1000032

4545

2500

67867868

76

21

2345

6347264

78675

89886757

435

3432423

234123

3434

89089

21

1252

562356

2342343

342455

78787

3333333

9999993

8789

123

13427777

4523523

56743

23423

6767312

342564

123123123

789976

4523523

1212

20002

1343414

2009

2346098

33129089

454544711

78787

3665364

Processed by
core 3

Processed by

SAP HANA

core 4

15

SAP HANA

30

SAP Business ByDesign

SAP HANA

SAP
(side-by-side)


SAP

SAP

SAP HANA IT
SAP HANA
-

SAP HANA,
www.experiencesaphana.com www.sap.com/hana.

J. Krueger, M. Grund, C. Tinnefeld, J. Schaffner, S. Mueller, and A. Zeier,

Enterprise Data Management in Mixed Workload Environments
(Hasso Plattner Institute for IT Systems Engineering, 2009)
http://ares.epic.hpi.uni-potsdam.de/apps/static/papers/2009
_JK_Enterprise_Data_Management_in_Mixed_Enviroments.pdf
H. Plattner, A Common Database Approach for OLTP and OLAP Using
an In-Memory Column Database (SIGMOD09, June 29July 2, 2009)
http://www.sigmod09.org/images/sigmod1ktp-plattner.pdf
H. Plattner, Enterprise Applications OLTP and OLAP Share One
Database Architecture (Hasso Plattner Institute for IT Systems
Engineering, 2010)
http://epic.hpi.uni-potsdam.de/pub/Home/InMemoryDataProcessing
2010/ShareOneDB.pdf
H. Plattner and A. Zeier, In-Memory Data Management: An Inflection Point
for Enterprise Applications (Springer-Verlag, April 2011)
Analyzing Business as It Happens: SAP In-Memory Appliance Software
(SAP HANA) runs on the Intel Xeon processor to generate superior,
real-time business intelligence (Intel|SAP, April 2011)
http://www.intel.com/en_US/Assets/PDF/whitepaper/mc_SAP_wp.pdf
D. Mettica and S. Lucas, Prepare for the Quantum Leap in Real-Time
Analytics. How in memory analytics is going to change everything about
your enterprise (IBM|SAP, June 2011)
http://www.sdn.SAP.com/irj/scn/go/portal/prtroot/docs/library/uuid
/d023f28a-6f9b-2e10-ec83-d84e518c3629?QuickLink=index

1. Source: Plattner, Hasso and Zeier, Alexander, In-Memory Data

Management: An Inflection Point for Enterprise Applications (SpringerVerlag, April 2011).
2. Ibid.
3. Restricted subset of C++ used internally at SAP.
4. Source: Analyzing Business as It Happens: SAP In-Memory Appliance
Software (SAP HANA) runs on the Intel Xeon processor to generate
superior, real-time business intelligence (Intel|SAP, April 2011).
5. Ibid.

SAP HANA

17

www.SAP.com/contactSAP

50 110 843 (12/01) 2011 SAP AG. All rights reserved.

SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign,
SAP BusinessObjects Explorer, StreamWork, SAP HANA, and other
SAP products and services mentioned herein as well as their respective
logos are trademarks or registered trademarks of SAP AG in Germany
and other countries.
Business Objects and the Business Objects logo, BusinessObjects,
Crystal Reports, Crystal Decisions, Web Intelligence, Xcelsius, and other
Business Objects products and services mentioned herein as well as their
respective logos are trademarks or registered trademarks of Business
Objects Software Ltd. Business Objects is an SAP company.
Sybase and Adaptive Server, iAnywhere, Sybase 365, SQL Anywhere, and
other Sybase products and services mentioned herein as well as their
respective logos are trademarks or registered trademarks of Sybase, Inc.
Sybase is an SAP company.
All other product and service names mentioned are the trademarks of
their respective companies. Data contained in this document serves
informational purposes only. National product specifications may vary.
These materials are subject to change without notice. These materials
are provided by SAP AG and its affiliated companies (SAP Group)
for informational purposes only, without representation or warranty of any
kind, and SAP Group shall not be liable for errors or omissions with respect
to the materials. The only warranties for SAP Group products and services
are those that are set forth in the express warranty statements
accompanying such products and services, if any. Nothing herein should
be construed as constituting an additional warranty.