The relational database model was first introduced by E.F. Codd of IBM in 1969. In the relational model, data are structured into tables (i.e., "relations") consisting of rows and columns. Each row contains a single record comprised of individual data elements (or "attributes") organized in columns containing elements of the same kind according to the rules defined for that column. Alternate database models include the network, hierarchical, flat file, and object-oriented models.
Video of the Day
The relational model structures data in a manner that avoids complexity. The table structure is an intuitive organization familiar to most users, particularly those who have worked with physical or software spreadsheets, check registers or other tabular data. Data are organized naturally within the model, simplifying the development and use of the database.
Ease of Data Retrieval
Under the relational model, accessing data in a database does not require navigating a rigid pathway through a tree or hierarchy. Users can query any table in the database, and combine related tables using special join functions to include relevant data contained in other tables in the results. Results can be filtered based on the content of any column, and on any number of columns, allowing users to easily retrieve meaningful results. Users can choose which columns to include in the results so that only relevant data are displayed.
Data integrity is an essential feature of the relational model. Strong data typing and validity checks ensure data fall within acceptable ranges, and required data are present. Referential integrity among tables prevents records from becoming incomplete or orphaned. Data integrity helps to ensure accuracy and consistency of the data.
The relational database model is naturally scalable and extensible, providing a flexible structure to meet changing requirements and increasing amounts of data. The relational model permits changes to a database structure to be implemented easily without impacting the data or the rest of the database. The database analyst can quickly and easily add, remove, and modify tables and columns in an existing database to meet business requirements. There is theoretically no limit on the number of rows, columns or tables. In reality, growth and change are limited by the relational database management system and physical computing hardware, and changes may impact external applications designed for a specific database structure.
A systematic methodology exists for ensuring a relational database design is free of anomalies that may impact the integrity and accuracy of the database. "Database normalization" provides a set of rules, qualities and objectives for the design and review of a database structure. Normalization objectives are described in levels called "normal forms." Each level of normalization must be completed before progressing to the next level. A database design is generally considered normalized when it meets the requirements of the third normal form. Normalization provides designers with confidence the database design is robust and dependable.