Storing Cubes and Aggregations

Cube data and aggregations can be stored in a variety of modes. Aggregations are precalculated summaries of data that provide the mechanism for rapid query response in OLAP systems.

Cubes can require substantial storage to contain the data and precalculated summary information in multidimensional structures. One factor that affects storage requirements is sparsity (the amount of empty cells in a cube). For example, if one dimension contains sales representatives and another dimension contains regions, cells at the intersection of the Northeast sales representative and the Southwest region would probably be empty.

Microsoft® SQL Server™ OLAP Services addresses cube storage requirements with a variety of techniques:

• Storage is not allocated for empty cells.
• Data compression is employed.
• A sophisticated algorithm designs efficient summary aggregations to minimize storage without sacrificing speed.

Storage options enable you to select appropriate storage modes and locations for cube data. You can create an OLAP storage strategy tailored to your needs.

In this section you can read about:

• Storage Modes (MOLAP, ROLAP, HOLAP)
• Aggregations