What is an OLAP Cube?

What is an OLAP Cube?


OLAP achieves the multidimensional functionality by using a structure called a cube. A cube is a collection of numeric data organized by arrays of discrete identifiers. Cubes were created to give a logical structure to the pivot table approach to analyzing large amounts of numerical data. It is a storage structure used by classic data warehousing products in place of many (often normalized) tables. Rather than using tables with rows and columns, cubes use dimensions and measures (or facts). Also, cubes will usually present data that is aggregated (usually summed), rather than presenting each individual item (or row). The structure of a cube is often described this way: cubes present a summarized, aggregated view of enterprise data, as opposed to normalized table sources that present detailed data. Cubes are populated with a read-only copy of source (or production) data. In some cases, cubes contain a complete copy of production data; in other cases, cubes contain subsets of source data. The data is moved from source systems to the destination cubes via extract, transform, and load (ETL) processes.


Creating an OLAP Cube using BIDS, you need to perform these steps:


• Create a new Analysis Services project

• Define a data source                 

• Define a data source view

• Invoke the Cube Wizard

Next Recommended Reading What are SQL Cubes?