Partition Wizard F1 Help (Analysis Services - Multidimensional Data)

You can use the Partition Wizard to define partitions for a measure group in a cube. By default, a single partition is defined for each measure group in a cube. Access and processing performance, however, can degrade for large partitions. By creating multiple partitions, each containing a portion of the data for a measure group, you can improve the access and processing performance for that measure group. For more information about partitions, see Partitions (Analysis Services - Multidimensional Data). You can start Partition Wizard from Cube Designer .

Warning

It is possible to create partitions that may contain incorrect data by including duplicate rows in the Specify Source Information or Restrict Rows pages. For more information, see Multidimensional Model Partition Management.

The Partition Wizard guides you through the following steps:

  • Selecting the data source view for the partition.

  • Creating a filter for the records included in the partition.

  • Setting the processing and storage locations.

  • Naming and saving the partition.

In This Section

See Also

Reference

Analysis Services Wizards (Multidimensional Data)

Concepts

Partitions (Analysis Services - Multidimensional Data)

Multidimensional Model Partition Management