Data Analysis and Normalisation
The basic concept of a relational database system is that data is only held once, ie there is no duplication, and the data can be accessed by many users at the same time for many purposes.
A ready-made, tailored database is not normally purchased off the shelf. You can employ the services of an external company to develop a tailored database for you. Not only is this expensive but the external company still needs to go through the processes as described here.
You can purchase the software to perform such functions as creating the tables, adding and amending data, creating reports, managing security, etc but the database itself still needs to be designed.
It is worth noting at this stage that the database, and the software to manage it, is collectively called a Database Management System. If the database conforms to the rules of a relational database, as originally defined by Dr E F Codd, then this is called a Relational Database Management System (RDBMS).
In this section we will examine briefly the process of data analysis and an introduction to a methodology to carry out a design, based on that analysis, known as normalisation.
When designing a database, it is important to ensure that it will support the organisation's needs both now and in the future. Furthermore, it should result in a design that avoids the pitfalls as described in the earlier sections.
A database designer examines in detail the data requirements of the system under development. The process involves analysing existing data files, screen layouts and reports, interviewing key staff, documenting the findings, etc. However, this is quite an extensive process and is outside the scope of this Unit.
The deliverables from this data collection exercise allow the Database Designer to use a data analysis technique to design the database.
Next: Data Analysis Techniques