Sie sind auf Seite 1von 22

0011 0010 1010 1101 0001 0100 1011

INTELLIGENT NAMING SYSTEM AN ALTERNATIVE NAMING SERVICE

Presented by: Shweta Mittal 6011138 Ram Lal Anand College (University of Delhi)

PAPER NO.:607
NAME : 0011 0010 1010 1101 0001 0100 1011 SHWETA MITTAL PROJECT REPORT TITLE : INTELLIGENT NAMING SYSTEMAN ALTERNATIVE NAMING SERVICE AREA OF PROJECT WORK : NETWORKS ROLL NO. : 6011138 NAME OF SUPERVISOR : MRS. NEELU MALIK NAME OF COLLEGE : RAM LAL ANAND COLLEGE ADDRESS OF COLLEGE : BENITO JUAREZ ROAD, NEW DELHI-110021. RESEARCH PAPER DETAIL: PAPER TITLE : INTELLIGENT NAMING SYSTEM: AN ALTERNATIVE FOR ENTERPRISE NAMING MANAGEMENT AUTHORS : LADDA PREECHAVEERAKUL, PATTARASINEE BHATTARAKOSOL JOURNAL : INTERNATIONAL JOURNAL OF COMPUTER SCIENCE & APPLICATIONS YEAR : 2006

2
2

ABSTRACT
0011 0010 1010 1101 0001 0100 1011
People use name in general to reference things easily. In addition, one name may refer to various types of things or objects . Name services have been developed to map logical names to physical resources. The structure of the current name services is hierarchical for scalability which limits one name mapping to many objects. This paper entitled INTELLIGENT NAMING SYSTEM: AN ALTERNATIVE FOR ENTERPRISE NAMING MANAGEMENT proposes the new naming system called the Intelligent Naming System. The technique called One Name Many Objects One Result (ONMOOR) is proposed to ensure that the result of the mapping is the intended object.

2
3

INTRODUCTION
0011 0010 1010 1101 0001 0100 1011

NAME SERVICE
Used to map logical names to any physical resources A distributed network database that contains key system information about all the systems on a network, so the systems can communicate with each other.

The name service has its own requirements:


Large Size Long Life High Availability Fault Isolation Tolerance of mistrust

2
4

0011 0010 1010 1101 0001 0100 1011

GLOBAL NAME SERVICE (GNS)


Used in an internetwork with a large naming database that stores resource

location and mail addressing. 0011 0010 1010 1101 0001 0100 1011

HANDLE SYSTEM
A general-purpose global name service that allows secured name resolution and
administration over networks.

NETWORK INFORMATION SERVICE

Sun Microsystems client-server directory service protocol for distributing system configuration data such as user and hostnames between computers on a computer network.

NETWORK INFORMATION SEVICE PLUS (NIS+)

Developed by Sun Microsystems to replace its older Network Information Service A name service for supporting management of basic network resources such as workstation addresses, security information, mail information etc.
6

NOVELL DIRECTORY SERVICE (NDS)


Centrally manages access to resources on multiple servers and computers within a given network. 0011 0010 1010 1101 0001 0100 1011 A database directory of all networked resources and services. Resources can include users, groups, printers, servers, applications, and so on.

DOMAIN NAME SYSTEM (DNS)

The essence of DNS is the invention of a hierarchical, domain-based naming scheme and a distributed database system for implementing this naming scheme. It is primarily used for mapping host names and e-mail destinations to IP addresses. Three major components: o Domain Name Space- specification for a tree structured name space and data associated with the names. o Name Servers- server programs which hold information about the domain trees structure and set information. o Resolvers- programs that extract information from name servers in response to client requests.

2
7

0011 0010 1010 1101 0001 0100 1011

2
8

LIMITATIONS:
Uniqueness

Character Support 0011 0010 Anonymity 1010 1101 0001 0100 1011 The table below shows the limitations for each name service:

This paper has mainly focussed on the limitation of uniqueness in the name of an organization that consists of many divisions and hence proposed the new naming system called Intelligent Naming System (INS).

2
9

0011 0010 1010 1101 0001 0100 1011

2
10

INTELLIGENT NAMING SYSTEM


The main purpose of designing INS is targeted to organizations with several faculties or divisions and each division is required to share the names which perform the same 0011 0010 1010 1101 0001 0100 1011 functions. Therefore, obtaining one accurate result from mapping one name to many objects is considered. We suppose that a university name Prototype University or PU consists of 4 faculties: Engineering, Medicine, Science, and Nursing. Each faculty consists of 3 offices: Academic Affairs (AA), Human Resources (HR), and Planning & Finance (PF). Though these offices have the same name, their faculties are different. Each office of every performs the same tasks.
Figure 1.

If we can refer to the name Human Resources for every faculty, it is easy to remember for everyone. Furthermore, using a single name for the same task in different faculty helps the organization maintains its unity.
11

faculty

STRUCTURE OF INS
0011 0010 1010 1101 the main0100 1011 of each name to be able to be shared Designed under 0001 consideration The structure employs the concept of sets and trees A global name space can be represented as labeled nodes. Each global name is a path in an inverted tree and may point to information of individual object, or a set of information of individual objects. The figure below illustrates the logical organization of the INS in the sense of a global name able to be shared:

Figure 2.

2
12

ACTIVE COMPONENTS
0011 0010 1010 1101 0001 0100 1011 3 significant components: INS is a client/server model. There are
1. INS CLIENT consists of an application and INS resolver (INSR). 2. INS SERVER contains mapping process (INSMP). an INS

3. INS COMMUNICATION is an INS message (INSM) which transmits queries and responses between client and server using TCP.

Figure 3.

The INSMP with filtering mechanism ensures that the result to a client would be the right unique object. This technique is called One Name Many Objects One Result (ONMOOR).

2
13

ONE NAME MANY OBJECTS ONE RESULT (ONMOOR) TECHNIQUE


0011 0010 1010 1101 0001 0100 1011
GLOBAL NAME PROPERTIES (GNP) The GNP is closely related to the filtering mechanism. Therefore, a global class (Gclass) and user class (Uclass) are defined. Gclass stores the information of each global object. Each global object contains 6 attributes: <Gname, Gobj, Glocality, GcrDate, Gcount, Gtype>.
Gname is a global name. Gobj is an object which relates to a global name. Glocality is a location that stores the object. GcrDate is a creation date of the object. Gcount is a total number of times the global name has been queried. Gtype is a type of the global name whether being a leaf or a node.

Uclass stores some information such as location.

2
14

INS MAPPING PROCESS (INSMP)


0011 0010 1010 1101 0001 0100 1011
The algorithm is described below: Let X be a local name server which is located nearby the client C. Let Y be a Cs location. Let T be a root name server which stores information of other name servers .

1. C queries a global name (g1:g2: :gk) to X. While X receives a message from C, X also stores Y. 2. X decomposes the query and forwards each subquery g1, g2, , gk to T. 3. T returns a name server, Ui, for each gi to X. 4. X queries g1, g2, , gk to name servers: U1, U2, , Uk, respectively. 5. Ui resolves gi by either returning data held on Ui or querying the set of name servers that has been delegated. Additionally, each Ui returns the sets of objects found (ansi) to X.
15

6. ANS is the intersection of all sets of objects (ans1, ans2, , ansk). If (ANS = {Obj1, Obj2, , Objn} then Call FilterProcess Else if ANS is a 0100 object 0011 0010 1010 1101 0001 unique1011 then X returns ANS to C Else X returns no result to C. Algorithm of FilterProcess

1. Use Quicksort algorithm to sort ANS ({Obj1, Obj2, , Objn }) based on counting factors: location and creation date, respectively. 2. X takes Y comparing to Obji in sorted ANS (sANS). 3. If one object (Obji) in sANS has a location as Y then Return Obji to C Else if no object in sANS has a location as Y then For each object in sANS Find Objj with a latest creation date Return Objj to C.

2
16

IMPLEMENTATION
The researchers have simulated client/server program as a prototype of INS and 0011 0010 1010 1101 0001 0100a1011 showed that INS can be worked properly. They have written an INS finder program used to resolve queries. Suppose that pu and sc.pu are name servers. The following examples show that the global names map the documented file(s): A client requests a query name that a local name server pu can resolve it.

A client requests a query name that a local name server pu cannot resolve it.

2
17

A client requests a query name with specific location.

0011 0010 1010 1101 0001 0100 1011

A client requests a query name with a combination of global names.

2
18

SUMMARISING THE FEATURES OF INS 0011 0010 1010 1101 0001 0100 1011
INS provides a facility for mapping a single name to many objects. The INS structure employs both sets and trees.

INS uses a client/server model. The INSMP with TCP packets guarantees that the data delivered between client and server will not be lost, especially the INS response data. INS uses UTF-8 encoding.

2
19

CONCLUSION
0011 0010 1010 1101 0001 0100 are implemented in a hierarchical structure. This The current name services 1011 structure contains a simple parent-child relationship which provides a unique name. The restriction assures that a name uniquely identifies a single leaf in the tree. Names at the leaf nodes of the tree represent individual objects. Thus, it limits names to be shared.
The new structure, on the other hand, employs both the concepts of sets and trees. Hence, names at the leaf nodes may not represent individual objects but a set of objects. The structure allows for one name to be shared. However, mechanisms exist so the correct object is returned to the client. The INSMP with a filtering mechanism based on user location or creation date has been proposed to find the required object.

2
20

REFERENCES
0011 0010 1010 1101 0001 0100An Alternative for Enterprise Naming Management, 1. Intelligent Naming System: 1011 by Ladda Preechaveerakul and Pattarasinee Bhattarakosol, Department of Mathematics, Faculty of Science, Chulalonkorn University, Bangkok. 2. Designing a Global Name Service, by B.W. Lampson, Digital Equipment Corporation, published in Proceedings 4th ACM Symposioum on Principles Distributed Computing, pp. 1-10 1986. 3. Development of the Domain Name System, by Paul V. Mockapetris and Kevin J. Dunlap, published in the Proceedings of SIGCOMM 88, Computer Communication Review Vol. 18, No. 4, August 1988, pp. 123133.) 4. Domain Names- Concepts and Facilities, by P. Mockapertis, in RFC1034, http://www.ietf.org/rfc/rfc1034.txt. 5. Domain Names- Implementation and Specification, by p. Mockapertis, in RFC1035, http://www.ietf.org./rfc/rfc1035.txt. 6. Handle System Overview, by S. Sun, Lannom and B. Boesch, in RFC3650, http://www.ietf.org/rfc/rfc3650.txt. 7. Computer Networks, Author: Andrew S. Tanenbaum.

2
21

0011 0010 1010 1101 0001 0100 1011

THANK YOU

2
22