Monday, August 6, 2007

Dimension Modeling

It is the type of model used by Data Warehouse Designers. It is the underlying data model used by most of the OLAP products today in the market. In this model Data is contained in two type of tables - Fact and Dimension tables.

DM - Fact Table

This contains the measurements or metrics or facts of business processes. In addition to the facts, the only other things are foreign keys for the dimension tables.

DM - Dimension Tables

It details in the dimension tables define the context of the measurements, such as who, where, how of a measure.

Dimension Attributes

These are various columns in the DIM Tables which gives additional info about the dimensional subject. This can have one or more hierarchical relationships and mostly used in report labels and query constraints.

Before building the Warehouse, decide what it is going to contain.

For Ex:

A typical Sales warehouse will have following dimensions.

Location
Time
Product

De-Normalizing the data in the OLTP tables into DIM Tables.

No comments: