Facts

A fact table is a database table that contains the measures and dimensions needed to relate a fact record to the dimension tables. To deliver fact data, fact builds are used. This topic is an overview of the facts in the sample catalog.

A measure is a value that can be analyzed, for example, number of defects. A dimension is a key component of a business for which you want to analyze data. The columns in a fact table correspond to the measures and dimensions. By default, referential integrity is enforced in the data warehouse. Therefore, fact table and the dimensions that are referenced in the fact table have referential integrity.

Fact data is delivered through fact builds. A typical fact build has the following transformation elements:

Fact table are of two kinds: trending tables and current snapshot tables. The trending fact table stores data every time an extract, transform, and load (ETL) job is run but current snapshot fact table stores data from only the last run ETL job. In the data warehouse, for each trending table, there is an associated current snapshot table. The trending table and current snapshot table have same structure. Using the data partition feature in IBM® Cognos® Data Manager, the same ETL build can be used to deliver data simultaneously to the trending and current snapshot fact table.


Feedback