Relational database

A relational database is a stored collection of digital information organized in table format to facilitate an easier search and retrieval process. Early computer databases were capable of storing data, but searching for that information was often slow and complex. In 1970, IBM computer scientist Edgar F. Codd developed the relational model of data, which organized information into tables and established searchable relationships among the data. A relational database aids in the search process but also allows tables to be expanded and new data to be added more easily. Data tables can range from a few categories to upward of hundreds of thousands for large collections of information. Relational databases are a standard means of accessing information and processing data transactions. The format is used to perform tasks from accessing bank and credit card accounts to making hotel reservations and online purchases.

Background

Prior to the growth of computer technology, most information was stored on paper and organized in various filing systems. In the eighteenth century, a system of cards with a series of punched holes was developed as an early form of automated data storage. The holes on the punch cards acted as “on-off” switches that carried a specific set of instructions. In 1890, American inventor Herman Hollerith developed a punch-card system to store data for the US Census Bureau. In the 1940s, many of the earliest computers used a variation of this system in the form of holes punched in strips of paper tape.

Prior to this, German inventors in the 1920s developed a way to store data on magnetic tape. Three decades later, researchers at the computer company International Business Machines (IBM) invented the hard drive, a storage technology that provided faster direct access to stored data than magnetic tape and made searching the data easier.

As the twentieth century progressed, technological advancements vastly improved computing capabilities. Compact discs (CDs), digital video discs (DVDs), and internet-based storage—often referred to as the “cloud“—allowed for an increasingly large amount of data to be stored. To manage such enormous amounts of information, computer scientists needed to develop a more efficient way to organize, search, and retrieve the data. Many organizations use cloud-hosted relational database services that provide scalable storage, automated backups, and reliable access to data.

Overview

The earliest computer databases were called flat databases, and they stored information in a long file consisting of a single row of text. Data values were separated by tabs, commas, or special characters such as a vertical bar (|). A database used to store bank records might have included first name, last name, address, and balance. For example, the following information might be stored in such a database:

John Doe, 123 Hill St., $1,000| Jane Doe, 42 Main St., $2,000| Tom Smith, 1 River St., $3,000|

Accessing the information required an advanced knowledge of computers and was often time-consuming. Users had to know where the information was stored beforehand and had to navigate through a string of text to find what they were looking for.

In 1970, Edgar F. Codd, a mathematician and computer scientist for IBM, developed a system that reorganized the stored data in tables consisting of rows and columns. Codd’s relational model of data, as he called it, allowed users to access specific data without first knowing its location. The system was also easier to use because users could retrieve information without navigating complex storage structures to find the data.

In relational databases, each column contains a data category, sometimes referred to as a field. Each row contains a specific record of data related to the category defined by the column. For instance, each column in a relational database for a bank defines a single topic—known as a field—such as customer name, account number, address, and balance. The rows contain the specific data associated with those fields—1234567, John Doe, 123 Hill St., $1,000, and so on. A bank employee using the database would search for relationships among the stored data. If a customer forgot the account number, an employee would search the table to view the number; a customer inquiry about available funds would prompt a search for data in the balance column. Relational databases can search for relationships among data on several levels. It can look for how one row or field relates to another row or field, or it can see how one row or field relates to many rows or fields.

A relational database also gives users the ability to limit or expand the amount of data in a table to meet specific needs. Smaller companies may include only ten tables in a database, while bigger companies may have one thousand or more. Large organizations may use relational databases containing thousands of tables to manage extensive collections of information. As new data or categories become available, relational databases can be easily modified to include the new information simply by adding additional rows or columns to the tables. Many relational database systems also support JavaScript Object Notation (JSON) and other semi-structured data formats in addition to traditional tables.

A programming language called structured query language (SQL) manages relational databases. SQL was first developed in the 1970s and later became the standard language for relational databases. The language is used for a variety of functions, including modifying database tables, adding columns or rows, and retrieving data from the tables. SQL commands used to change, insert, or delete data in tables use a programming language called data manipulation language (DML); commands used to create or modify database tables or structures use data definition language (DDL).

SQL commands are written as statements and typically include terms such as add, alter, create, delete, insert, select, truncate, and update. An SQL command to search a bank database for customers with account balances of more than $1,000 may look like this:

SELECT First_Name, Last_Name
FROM Balance
WHERE Balance > 1000

Relational databases remain widely used for applications that require accurate recordkeeping and reliable transaction processing, including banking, health care, government services, and electronic commerce.

Bibliography

“Amazon RDS.” Amazon Web Services, aws.amazon.com/rds/. Accessed 31 May 2026.

Ambler, Scott W. “Relational Databases 101: Looking at the Whole Picture.” Agile Data, www.agiledata.org/essays/relationalDatabases.html. Accessed 31 May 2026.

Captain, Fidel A. Six-Step Relational Database Design: A Step by Step Approach to Relational Database Design and Development, 2nd ed., CreateSpace Independent Publishing, 2013.

Codd, E. F. “A Relational Model of Data for Large Shared Data Banks.” Communications of the ACM, vol. 13, no. 6, 1970, pp. 377–87, doi:10.1145/362384.362685. Accessed 31 May 2026.

Harrington, Jan L. Relational Database Design and Implementation. 4th ed., Morgan Kaufmann, 2016.

“History of Data Storage Technology.” Zetta, www.scribd.com/document/774225467/2-History-of-Data-Storage-Technology. Accessed 31 May 2026.

“Relational Database: Definition, Examples, and More.” Coursera, 16 Oct. 2025, www.coursera.org/articles/relational-database. Accessed 31 May 2026.

“A Relational Database Overview.” Oracle, docs.oracle.com/javase/tutorial/jdbc/overview/database.html. Accessed 31 May 2026.

Taipalus, Toni. “Database Management System Performance Comparisons: A Systematic Literature Review.” Journal of Systems and Software, vol. 208, 2023, article 111872, doi:10.1016/j.jss.2023.111872. Accessed 31 May 2026.

“What Is a Relational Database? Example and Uses.” HowStuffWorks, 9 Oct. 2024, computer.howstuffworks.com/question599.htm. Accessed 31 May 2026.

“What Is a Relational Database?” IBM, www.ibm.com/think/topics/relational-databases. Accessed 31 May 2026.

Yasar, Kinza. “What Is Structured Query Language (SQL)?.” TechTarget, 27 Aug. 2024, searchsqlserver.techtarget.com/definition/SQL. Accessed 31 May 2026.