Design of a Dynamic Federated Database.

Date of Submission

December 1988

Date of Award

Winter 12-12-1989

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science


Theoretical Statistics and Mathematics Unit (TSMU-Bangalore)


Bagchi, Aditya (CSSC-Kolkata; ISI)

Abstract (Summary of the Work)

A federated dat.abase system (FDBS) is a collection of cooperating databases (DBSS) that are autonomous. Here we will develop a methodology for building such FDBS The a component. DBSS are integrated to varying degrees One of the significant aspects of FDBS is that component DBS can a a continue its local operations and at the same time participate in federation A key characteristic of the federation is a the cooperation among the independent systems. At present there controlled sharing is no ideal model for federation. To allow while preserving the autonomy of the component DBSS and execution of the existing applications FDBS continued an supports types of operations local and global. Global two operations involve data access using the FDBS and may involve data managed by multiple component DBSS. Local operations be submitted to a component DBS directly.The proposed strı i is dynamic in nature where a new database can easily be added to or withdrawn from the federation.We normally form single database for enterprise enterprise is reasonably self_cont an commercial, scientific,or other organization; e.s.. manufacturing company „bank,hospital,university etc. .The đatabase contains all the relevant data of the enterprise.In a relational system update the schema by adding we can new attribute(s) to relation or by forming relation a new according to the requirement.Update also includes deletion of attrbutecs) relation(s).In this way enterprise,however an complicated may be modelled by single database. The above approach has the following limitaions :1. A big enterprise ingeneral consists of a number of subenterprises. Though these subenterprises contain overlapping information (1.e. some relations fully are or partly common in the that they refer to of sense same set attributes Deach such subenterprise is large enough be to considered as a separate database.2. Each subenterprise has users who are interested only in that subenterprise and be interested other may not in subenterprises.3. So each such database modelling may need local autonomy for efficent maintenance and query processing. Subenterprises should act independent of each other so of error/crash in one subenterprise local operations in other subenterprises should not get affected. If4. that in the event we keep a single database for the whole enterprise,we cannot provide such facility.5. Data may be distributed among multiple databases. Benefits of data distribution such as increased availability as well as improved access time is evident.


ProQuest Collection ID:

Control Number


Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


This document is currently not available here.