Originally a RDBMS was a software dedicated to the management of attribute type alphanumeric data stored in tables and organized in structured databases. With standardized software attribute queries (SQL = Standard Query Language), a DBMS can query the tables to make joints or syntheses. GIS software all offer a RDBMS module responsible for the management of attribute data. More recently, DBMS have expanded the types of data they were able to manage. Now SDBMS (S for spatial) can include all geographic, attribute, raster and vector data, managed in an integrated and consistent manner. However, the classical model associating a table to a layer of entities is still dominant in simple systems.

The organization of data in a RDBMS must be very strict and formalized by a conceptual data model. The correspondence between two tables must be made through a single column (or field) in each related table. Generally, relationships use columns that serve as identifiers to make sure the relationship is stable over time and is based on many unique values.

See: Geographic Database