Data Warehouse

A data warehouse contains business information organized and stored in such a way that queries and reports can be run quickly with less impact on a transactional database.

Note: The terms "data mart" and "data warehouse" are sometimes used interchangeably. But generally, "data warehouse" indicates a large repository of business information at an organization and "data mart" indicates a smaller or specialized repository that accommodates a department or division of an organization. A data warehouse feeds a data mart, from which queries and reports are created.

Data warehouses are usually maintained separately from an organization's day-to-day transactional database system. As such, the database for a data warehouse is periodically updated with data from the transactional database. Ideally updates occur when traffic to the transactional database is low, such as during non-working hours. This separation ensures that queries and reports run on the data warehouse do not affect the operation of the transactional database.

Data warehouses are also structured differently than transactional databases. While transactional databases are designed to perform well for a variety of functions such as small queries and updating records, data warehouses are designed to perform well for reports and queries that involve a large number of records. There are many ways to structure a data warehouse database. One such structure, the star schema, is discussed in Star Schema.