Database Design for Smarties: Using UML for Data Modeling
Learn UML techniques for object-oriented database design.
TABLE OF CONTENTS Database Design for Smarties: Using UML for Data Modeling
Chapter 12: Designing an Object-Relational Database Schema
Ex umbris et imaginibus in veritatem. "From shadows and types to the reality."
Cardinal Newman's motto
Overview
Object-relational databases extend the capabilities of the relational database toward representing objects. Although there are no standards yet in this arena, a clear picture emerges from the competitive landscape and from the ongoing standards efforts of SQL3.
So What's New?
ORDBMS products are also known as "universal" servers or database managers, presumably because they can handle any kind of data, or possibly because vendor marketing would like their customers to adopt the new technology universally (very small joke). ORDBMS products add several basic things to a relational system:
-
Large object data types
-
Extended data types
-
Multivalued data types
-
Object identity
-
Extensions that aid in using the above new features
Features
First, ORDBMS products add extra types to the type system to represent objects of arbitrary complexity: BLOBs, CLOBs, FLOBs, and so on. LOB stands for "large object," and these data types represent objects by giving you access to the computer representation of the object as SQL data. On reflection, though, this makes it sound like more than it is. In reality, LOB data types let you put some arbitrary set of bits or characters into a database and that's it. The more advanced databases let you get the bits or characters out again, but it's up to you to make sense of them. The ANSI SQL3 standard specifies three kinds of large object: the CHARACTER LARGE OBJECT (CLOB), NATIONAL CHARACTER LARGE OBJECT (NCLOB), and...
Copyright Academic Press 1999 under license agreement with Books24x7
