1 By Paul Murray Claire McQuade Kashif Rafiq David Miller.

1

ByPaul Murray

Claire McQuadeKashif RafiqDavid Miller

2

Introduction Federated Databases (FDBSs) Key areas

Characteristics of DBS/FDBS Architecture Views

3

Characteristics - of DBS

• Distribution

• Autonomy

• Heterogeneity

4

Characteristics - Autonomy

Autonomy

Logically Integrated Multiple DBMS

Federated DBMS

Multidatabase System

low High

Transaction Control

Query Processing

Distribution of ControlDegree to which individual DBMS can operate independently

5

Characteristics-Distribution

Deals with data

Single DBS Many DBSs in a local area network Many DBSs in a wide area network

Distribution

Single DBS Multiple Sites

Local Distributed

6

Characteristics-Distribution

Data and the Federated Database System (FDS) Databases may be on the same computer Databases may be geographically

separate Systems must be able to communicate

Benefits of distribution Improved access times Improved availability Improved reliability

7

Characteristics -Heterogeneity Data models

Structures Constraints Query languages

8

Characteristics of FDBS Collection of cooperating but autonomous

component database systems (DBSs)

Component DBSs integrated to various degrees

FDBMS Software that provides controlled and

coordinated manipulation of the component DBSs

No centralized control

9

Types of FDBS

Multidatabase System

Loosely coupled

Federated DBSNonfederated DBS

Single Federation Multiple Federation

Tightly coupled

10

Components of FDBS

Federated Database System

System 1 DBMS (centralised)

System 2 DBMS (distributed)

System N DBMS (FDBS)

11

Architecture Processor Coupling Schemas

12

Architecture - Processor Transforming Processors Filtering Processors Constructing Processors Accessing Processors

13

Transforming Processors

SCHEMA A

TRANSFORMING PROCESSOR

SCHEMA B

Command TranslationSchema Translation

14

Filtering Processors

SUBSET OF SCHEMA A

FILTERING PROCESSOR

SCHEMA A

Control CommandsControl Data

15

Constructing Processors

SCHEMA A

CONSTRUCTING PROCESSOR

SCHEMA B SCHEMA C

16

Accessing Processors

COMMANDS

ACCESSING PROCESSOR DATA

DATABASE

17

Architecture - Coupling Loosely Coupled FDBS Tightly Coupled FDBS

18

Architecture - Schema Schema Models

Five layer model Seven layer model Schema evolution

19

Schema - Five-Level

Component DBS Component DBS

Local Schema

Component Schema

Export Schema Export Schema

Federated

Schema

ExternalSchema

Local Schema

Component Schema

Export Schema

Federated

Schema

ExternalSchema

External Schema

20

Schema-Five-Level Processor

Component DBS

Local Schema

Component Schema

Export Schema

Federated

Schema

External Schema

Transforming Processor

Filtering Processor

Constructing Processor

Filtering Processor

Component DBS

Local Schema

Component Schema

Export Schema

Federated

Schema

External Schema

Transforming Processor

Filtering Processor

Constructing Processor

Filtering Processor

21

Schema - Evolution One of the problems found with

federated databases is the cumulative schema evolution rate A client application cannot be tied to

a specific integrated view of the federation, as all it takes is one member of the federation to change its schema without ensuring backward compatibility to cause serious problems

22

Semantic - Integrity Constraints The federated database schema should

correctly reflect the semantics of the component database schemata of which it is composed

Since the semantics of a database schema is also determined by a set of semantic integrity constraints, a correct schema integration has to deal with integrity constraints existing in the different component database schemata

23

Architecture - Views Selection Projection Join Real

24

Conclusion