SIMPLE DATABASES

Simple Databases



Simple Databases

Introduction

A database is a collection of interrelated data of different types. The term conveys much more than the older term file (q.v.). Unfortunately, “database” is still all too frequently used when all that is implied is a conventional file. The difference between a database and a file, in terms used prior to the advent of data processing, is analogous to the difference between a thoroughly cross-referenced set of files in cabinets in a library or an office and a single file in one cabinet which is not cross-referenced with any other file. The important difference between a computerized database and a thoroughly cross-referenced set of manual files is that the database must be stored on a direct access storage device in order for the computer's central processing unit (CPU) to be able to use the cross-references (Date, 1994).

There are different ways of talking about databases, and consequently different approaches and different sets of terminology. The most widely used approach is called relational, and a more recent one is referred to as object-oriented database management. Furthermore, some of the ideas which stem from the object-oriented approach have been included in relational systems, giving rise to the term object-relational databases.

Tables, Columns, and Constraints

A conventional file, definable in Cobol since the early 1960s, is regarded as a collection of records of the same or possibly different types. In the relational approach, a file is referred to as a table containing columns and tuples (rows). A row in a table is essentially the same as a record (q.v.) in a file. A column in a table is what in Cobol is called an item (or, more correctly, an elementary item).

Constraints

The difference between the database and Cobol approach comes in the use of constraints in a database. There are different kinds ...