Component Database Systems
Nine individual overviews of how "componentization" is changing the face of database technology.
TABLE OF CONTENTS Component Database Systems
Section 2.2: Single-Site Component DBMSs
2.2 Single-Site Component DBMSs
In this section we give an overview of how a single-site ORDBMS works. In most cases, vendors implement their distributed ORDBMS products by augmenting the functionality of a single-site system. This strategy is motivated more by commercial than technical considerations. The market for distributed ORDBMSs is a subset of the broader market for single-site DBMS software. And as we shall see, much of the code in a single-site system is useful in the distributed system.
2.2.1 ORDBMS Abstract Data Model
As mentioned in the introduction, the most important difference between traditional DBMSs and ORDBMSs is that ORDBMSs manage both persistent data and logic implementing operations over to this data within an abstract data model. This permits developers to use declarative expressions to organize and manipulate the objects for which their information system is responsible. In other words, instead of creating programs wherein every user-level operation is hand-coded and compiled into an executable form, the ORDBMS encourages more flexible system architectures. With an ORDBMS, complex business questions and changes to the information system can be handled at runtime by building appropriate query expressions.
Example 1
Figure 2.3 presents two SQL fragments illustrating this. Figure 2.3(a) creates a table to store facts about subscribers in our cellular network system, and the query in Figure 2.3(b) records the fact that a call is beginning.
Figure 2.3: Query data definition language (DDL) and data manipulation language (DML) examples for cellular Operational Support Systems (OSS) application.
Readers acquainted with relational DBMSs...
Copyright Academic Press 2001 under license agreement with Books24x7
