Definition of Database

Many non technical people in IT wonder, what is a database? Well one a common definition of database is that it is the a collection and logical arrangement of small pieces of information called data which are related in some way to each another. It is usually utilised during the process of software development as well as during various stages of software testing.

3 Essential Properties of a Database

1. It is a collection of data which are small pieces of information.
2. The data stored in a database are related to each other
3. The collection is done in a logical manner

According to the above definition of database, a simple phone book, a record of household expenditure and the sales data from a shop are in theory at least, all a form of a database. The common factor amongst all of these examples, is that the data is arranged in the form of a table with rows and columns, as you would find in an excel sheet.

Another definition of a database is that it is an organised mechanism which helps in the storage, management and retrieval of information. Think of it as being similar to the way in which your memory works

Different Kinds of Database Models

In order to understand, what is database? we have explained a number of the various models which have been created. These have then evolved into various different types of databases when used in the real world.

Please note that all databases may use any of these models for data arrangement and management. The kind of model also determines the various operations that can be performed on the data stored in a database model.

1.0 Flat Model

A flat model consists of a two-dimensional array of rows and columns where the column data is of similar nature; be it date and time, integers, character data or floating point number. The most widely known example of a flat model is an excel spreadsheet.

2.0 Hierarchical Model

A hierarchical model stores data into a reversed tree-like or parent-child structure where each record has one single upward link. There is one table which acts as a root for all other tables, which in turn may further act as branches from which some more tables originate. This model was used in Information Management System (IMS) by IBM.

3.0 Network Model

This is an improvement over the hierarchical model where a record can have more than one upward linking. Instead of a regular parent-child model where a child can have only one parent, the network model allows linking to more than one parent.

4.0 Relational Model

This model was designed to overcome the shortcomings of previous versions. In this, the data is stored in various tables which are given specific names for the purpose of identification. Also there are unique identity field called the primary key in each table which is used to relate a table with others. These primary key fields are also used in retrieving data from two or more different tables at the same time. The Relational model is used in the latest database management systems such as SQL server and Oracle.

5.0 Dimensional Model

The Dimensional model is used for representing data in data warehouses. The database under this model consists of a large table that uses the dimensions of a cube to represent key data entities. The data stored in this model has very fast response time for drilling down or rolling up the table.

6.0 Objectional Model

The Objectional model is the object-oriented approach to database technology. This model ensures that the database uses the same type-system as its application program. This helps in avoiding any kind of mismatch during the process of software development when the information in the database representation is being converted into application representation. It has found its application in engineering databases and molecular biology databases.

Definition of Database - Tip

When your business analyst is writing up the business requirements documentation (BRD) and your solution architect the software requirements specification (SRS) make sure you check that they are correctly documenting firstly what database functionality they need, and secondly exactly what physical changes will be required.

---