Overview RDBMS-ORDBMS-

OODBMS

1
 Database Models – Transition
Hierarchical Data Model

Network Data Model

Relational Data Model

ER Data Model

Semantic Data Model

Object-Relational DM Object-Oriented DM
2
 Main Issues

ƒ Relational DBMS Limitations

ƒ What is an OODBMS?
ƒ Advantages and Disadvantages of OODBMS
ƒ What is an ORDBMS?
ƒ What is SQL3?
ƒ Comparison of OODBMS and ORDBMS
ƒ When to use an OODBMS
ƒ When to use an ORDBMS
3
 Relational DBMS Limitations
ƒ Semantic overloading.

ƒ Poor representation of ‘real world’ entities

ƒ Poor support for integrity & business constraints.

ƒ Homogeneous data structure.

ƒ Limited operations.

ƒ Difficulty handling recursive queries.

ƒ Difficulty with ‘Long Transactions’.

4
Relational DBMS Limitations
ƒ Normalisation (Normal Forms and FDs) sometimes lead to
relations which do not exist, or correspond, to entities in the
real world. This compounds on the ‘join’ feature of query
processing

ƒ The many to many relationship is difficult to express.

ƒ The RDBMS has domains, keys, multi-valued and join

dependencies

5
 OODBMS Architecture

OBJECT ORIENTED DATABASE MANAGEMENT SYSTEM

OOPL DBMS
ƒComplex Objects Persistence
ƒObject Identity Disc Management
ƒMethods & Messages Data Sharing
ƒInheritance Reliability
ƒPolymorphism Security
ƒExtensibility Ad Hoc Querying
ƒComputational Completeness

6
 OODBMS Main Features
ƒ Complex objects
ƒ Object Identity
ƒ Encapsulation
ƒ Classes
ƒ Inheritance
ƒ Overriding and late-binding
ƒ Extensibility
ƒ Computational completeness
ƒ Persistence
ƒ Concurrency
ƒ Recovery
ƒ Ad-hoc querying

7
 Storage-Levels

RDBMS Storage ODBMS Storage

Main or virtual
memory Main or virtual
memory
Transforming and
type checking
SQL
Secondary
Secondary storage
storage

8
 OODBMS Advantages

ƒ Enriched modelling capabilities

ƒ Extensibility

ƒ Support for schema evolution.

ƒ Applicable for advanced database applications

ƒ Improved performance.

9
 OODBMS Disadvantages

ƒ Lack of a universal data model ?

ƒ Ad-hoc querying compromises encapsulation.

ƒ Locking at object-level impacts performance

ƒ Complexity

ƒ Lack of support for views

ƒ Lack of support for security

10
 Then What is an ORDBMS ?

ƒAn Object-Relational database adds features associated with object-

oriented systems to a RDBMS
OR
ƒExtend the relational data model by including object orientation
and constructs to deal with added data types

11
Object-Relational DBMS Features

OODBS support noted by RDBMS vendors include

ƒ User-extensible type system
ƒ Encapsulation
ƒ Inheritance
ƒ Polymorphism
ƒ Dynamic binding of methods
ƒ Complex objects including first normal form objects
ƒ Object Identity

12
ORDBMS Drawbacks
Disadvantages of ORDBMS
ƒ Complexity
ƒ Increased costs
ƒ Unclear if the ORDBMS will actually combine
relationships and encapsulated objects to correctly and
completely mirror the ‘real world
ƒ Provision of a language(s) which will front end to SQL
and will provide a migration path for existing SQL
users

13
 ORDBMS-SQL3
SQL3 is a superset of SQL/92, in that it supports all of the
constructs supported by that standard, as well as adding new
ones of its own.
New Types
ƒ Extended Base Types.
ƒ Row Types.
ƒ User-Defined Types.
ƒ User-Defined Routines.
ƒ Sub-Types and Super-Types.
ƒ Sub-Tables and Super-Tables.
ƒ Reference Types and Object Identity.
ƒ Collection Types.
14
 OODBMS or ORDBMS ?

ƒOODBMS put more emphasis on the role of the client side

This can improve long, process intensive, transactions.

ƒORDBMS SQL is still the language for data definition,

manipulation and query

ƒOODBMS have been optimised to directly support object-

oriented applications and specific OO languages

ƒORDBMS are supported by most of the ‘database vendors’

in the DBMS market place

15
 OODBMS or ORDBMS ?

ƒORDBMS Most third-party database tools are written for

the relational model and will therefore be compatible with
SQL3

ƒORDBMS search, access and manipulate complex data types

in the database with standard (SQL3 ?), without breaking the
rules of the relational data model

ƒOODBMS The ODMG standard group’s OQL is now the

standard query language amongst OODBMS vendors

16
 OODBMS or ORDBMS ?

When to use an ODBMS?

ƒIn applications that generally retrieve relatively few

(generally physically large) highly complex objects and
work on them for long periods of time.

When to use an ORDBMS?

ƒIn applications that process a large number of short-

lived (generally ad-hoc query) transactions on data
items that can be complex in structure

17
 Concluding Remarks

ƒ OODBMS:
Abandon SQL (use an OO language instead)

ƒ ORDBMS:
Extend SQL (with OO features)

18
 Concluding Remarks

ƒObject-oriented capabilities have in effect turned the relational

model and the process of normalization on its head.

ƒVarious nested objects and/ or references can be queried without

doing a join.

ƒA shift from the relational to the object-oriented model is taking

place ?

19