Star Schema

Data warehouses are organized according to a schema. Blackbaud Data Warehouse uses a modified star schema. Star schemas are a set of fact tables that reference dimension tables. Fact tables contain primary data and dimension tables contain secondary, attribute-like data. For example, the Revenue star has a fact table called FACT_REVENUE and several dimension tables including DIM_PURPOSELEVEL.

In the illustration, there are connections between dimensions and fact tables. But when you look at a fact table in a BBDW database, you'll notice there are no foreign keys. Relationships between fact tables and dimensions are catalogued through comments on the tables and columns in a BBDW database. Further, BBDW database tables are not arranged in a pure schema. Views that abstract BBDW database tables create the star schema.

Unfortunately, to explore a star at the BBDW database level, you have to do some investigating. However, when you look at the data model from the data source view level in Business Intelligence Development Studio through an Analysis Services connection, you can see diagrams about how the structures fit together. For more information, see Analysis Services and Business Intelligence Development Studio (Briefly).

When you extend the BBDW database, you add tables and the comments that indicate relationships between tables through database revisions. For more information see BBDW Database Revisions Extensions and Revisions Spec: Map Source to Target.