The concept of a relational database was first established in 1969 by Edgar Frank Codd, a British mathematician and former Royal Air Force pilot working in an IBM laboratory in San Jose, California. Because doing calculations on computers cost hundreds of dollars per minute, time was at a premium. Codd's insights into storage, retrieval and analysis of data helped overcome cost barriers and build the database industry into a multibillion-dollar business.
The main feature of any relational database is its primary key. The primary key is a unique number or identifier assigned to each record that travels across different tables. A good example of a primary key is a Social Security number. The primary key makes each record unique, and it allows data to be kept in more than one table. Each table within a relational database will have a field for the primary key.
A relational database has more than one table with data. The tables are joined by the primary key. The key can serve to join tables together in many different relationships, such as a one-to-one relationship, where each table contains one unique record; a one-to-many relationship, where one table might contain records about an individual and the other might have records about all that individual's financial transactions; or a many-to-one relationship, where one table might contain a list of individuals and the other might have data about the type of customers represented by those individuals.
All databases will allow their owners to create, read, update and delete (CRUD). This is generally done through various forms of Structured Query Language, designed in 1974 by IBM, and based upon principles of relational algebra. The language also allows most users to query and manipulate data, as well as protect it from unwanted updates, deletions or other potential errors.