A cube is a multidimensional structure that contains dimensions and measures. Dimensions define the structure of the cube, while measures provide the numerical values of interest to the end user. Cell positions in the cube are defined by the intersection of dimension members, and the measure values are aggregated to provide the values in the cells.
The following topics provide a basic conceptual overview of cubes.
|Introduction to Cubes||Introduces the concepts and behavior of cubes, including discussions on the structure, storage, and processing of cubes.|
|Cube Structure||Details the components of a cube, such as measures and dimensions, and discusses the importance of the cube schema.|
|Cube Storage||Describes the various techniques and storage modes, including the various uses of partitions, linked cubes, distributed partitioned cubes, and real-time cubes.|
|Cube Processing||Describes the nuances of processing cubes, including information on real-time OLAP.|
|Cube Varieties||Provides a brief overview of the varieties of cubes available in Microsoft® SQL Server™ 2000 Analysis Services.|
|Regular Cubes||Discusses regular cubes, which are based on tables. Regular cubes have their own aggregations, which utilize storage space. Regular cubes are distinct from linked cubes, virtual cubes, and local cubes.|
|Linked Cubes||Discusses linked cubes, which are based on regular cubes defined and stored on another Analysis server. A linked cube uses the aggregations of the cube on which it is based; therefore, it requires no storage space for aggregations.|
|Distributed Partitioned Cubes||Discusses distributed partitioned cubes, which are regular cubes with partitions stored on another Analysis server. A distributed partitioned cube uses aggregations that can be based on both local and remote Analysis servers, and thus can distribute processing across multiple Analysis servers.|
|Virtual Cubes||Discusses virtual cubes, which are logical cubes based on one or more regular cubes or linked cubes. Virtual cubes use the aggregations of their component regular cubes and the cubes on which their component linked cubes are based. Thus, virtual cubes require no storage space for aggregations.|
|Local Cubes||Discusses local cubes, which are contained in portable files and can be browsed without a connection to an Analysis server. Local cubes are based on tables. They do not have aggregations.|
|Real-Time Cubes||Discusses real-time cubes, which are regular cubes with dimensions and/or partitions enabled for real-time OLAP. The dimensions and/or partitions enabled for real-time OLAP and used by real-time cubes require no storage space for aggregations.|
|Write-Enabled Cubes||Discusses write-enabled cubes, in which authorized end users can update the cube's cell values.|