Start of changeIBM FileNet P8, Version 5.2.1            

Fact and dimension tables

For purposes of IBM® Cognos® Business Intelligence integration with background searches, a background search produces two main types of result tables: fact tables and dimension tables.

Dimension tables have a primary key with a corresponding foreign key value that is stored in the fact table. The relationship between the primary key and foreign key does not have to be enforced because Cognos reporting tools can assign the relationship through the modeling definitions.

Fact tables

The fact table is the table in which the background search query stores the search results. Each cell of an OLAP cube holds a measure (such as sales or profits) that is derived from a fact table.

You might rely on Cognos Business Intelligence post-processing to perform any needed aggregation of fact table data. An optional alternative for performing data aggregation is the use of aggregate functions in the search query. Some examples of aggregate functions are COUNT, MAX, MIN, and SUM.

Dimension tables
The summarization of OLAP factual data occurs along dimensions (such as time or region) that are derived from dimension tables. Cognos Business Intelligence can analyze multidimensional data (measures by dimensions) interactively from multiple perspectives, including the following ones:
  • Consolidation (roll-up)
  • Drill-down
  • Slicing and dicing
  • Pivoting

Dimension tables can be created or populated in the following ways:
  • Manually. You can create dimension tables manually.
  • Automatically. You can use custom search functions to populate dimension tables. The handlers that you define for custom search functions enable the generation of data based on the post-processing of each row. Some custom search functions are installed by the Reporting Enablement AddOn add-on. For example, the CmRpt::FormatDate() function can extract and reformat date portions from an input date property value. (This add-on also contains sample background search queries as well as supporting objects to facilitate the generation of reporting data.) For more information, see Custom search function example.


Last updated: March 2016
p8pcc452.htm

© Copyright IBM Corporation 2016.
End of change