A fact table is a logical container for the relational tables defining the data values (measures) for each dimension intersection in an OLAP model.
The type of Essbase database you create determines which source tables to include in the fact table. See Defining User Needs.
When selecting source tables for the fact table, consider data distribution in tables and how the data relates to dimensions in the Essbase database.
Select tables containing measure data that Essbase users want to see. For example, if the source database contains ORDERS and ORDERDETAILS tables, include both tables.
Select tables that represent dimensions (data categories) on which users want to analyze data. For example, if users need to analyze product sales by time and sales channel, select tables containing key columns for these dimensions.
You may improve performance by creating aggregate data in tables instead of in views. See Creating Views, Tables, and User-Defined Tables. See Deciding to Create Staging Areas.
If the Essbase users need to only aggregate data, then include tables or views of aggregated data. You can add more detail as you build the OLAP model.
If your data source is a relational database, you can improve performance by setting up primary and foreign keys, and joining the tables that form the fact table. See Adding Columns to Tables. |