The purpose of a Data Warehouse (DW) is to consolidate and organize data so it can be analyzed and used to…

DW usually contains historical data, often collected from a variety of sources such as OLAP systems, legacy systems, data marts. A DW combines this data, cleanses it for accuracy and consistency, and organizes it for ease and efficiency of querying.

In large data warehousing applications, data is often segmented into specialized components, called data mining cubes, that address individual components of the organization.

Referential integrity must be maintained when DW data is added, or deleted. Loss of referential integrity can result in errors during cube processing, fact table records being bypassed, or inaccurate OLAP information.

Cube information available online to client applications cannot be affected when data is added to the DW due to interaction between the data and cube partitions. Every DW defines partition filters that avoid this problem.

ROLAP's multidimensional data model answers to a query into historical data and never leads to subsequent queries as the analyst searches for answers or explores possibilities. ROLAP systems provide the speed and flexibility to support the analyst in real time.