Partitioning allows tables and indexes to be subdivided into smaller pieces, and each small piece (eg., each subset of the overall data) is called a partition.
Partitions are typically created based on a date field.
This article covers how to create a partition schema using the Jet Data Manager, and how to use that partition schema to partition a table.
1. Right-click the table that you would like to setup table partitioning on, and select Advanced >> Advanced Settings
2. Click the plus (+) sign to create a partitioning template
3. Give the template a descriptive name, and select among the options available. In this example, we will partition using the Month (YYYYMM) operation, the use of which will become apparent in Step 4
4. Complete the partitioning setup by providing values for the following three fields:
a. Template: Pick the template you created in step #3 from the drop-down list
b. Field: Specify a date field from the table you are partitioning. In this example, I selected Posting Date
c. Time Table: The time table chosen in this list must be a time table (such as the Date table in the data warehouse for all CubeStore project versions up to v2.6)
5. Click OK , and deploy & execute your table. Notice the creation of a new partition field in the format of YYYYMM inside the table, as well as a Partition Templates folder inside the Jet Data Manager. Partition templates can be re-used in other tables
6. Once that completes you'll need to Edit the cube using the table you setup the partition on and select the Incremental Load option.
When the cube gets executed again, it will utilize the partition.