AQA Information and Communication Technology/ICT5/Database Management Concepts
Early databases were flat-file, meaning they were small, had lots of duplicated data and could not relate data to each other. As a result of data duplication, data can be inconsistent in separate parts of the database. Flat-files also tended to be unique to a particular program, making data portability difficult.
A database management system (DBMS) was conceived to help alleviate these shortcomings. This model gives three properties:
- Entities - An area of interest about which the data may be held
- Attributes - Properties of entities
- Relationships - A link between entities
There are different types of relationships:
- One-to-one - where an entity has a single link to another entity
- One-to-many - where an entity may link to many other entities
- Many-to-many - lots of one-to-ones
By using relationships, you can link data and have queries that will update linked data at the same time, saving time and increasing the accuracy of data. The Wikipedia has more information about DBMS's.
The role of a database administrator (DBA) is to manage and maintain the DBMS for most efficient usage. The DBA maintains access rights, relationships and updates the database of the design if need be.
Data normalisation is the process of increasing database efficiency by reducing duplication of records and producing the best possible database design. A normalised database is consistent and flexible.
Data consistency is ensuring that the data is consistent across all occurrences of it (this is made simple using relationships).
Data integrity is the quality of the data held in the database.
Data redundancy is where data is unnecessarily duplicated.
Data independence is where data should be independent of any other changes in the database.