Microstrategy Analytics

Considerations for the use of star schemas with MicroStrategy SQL Generation The primary characteristic of star schema is its use of dimension tables rather than single-attribute lookup tables. For example, a Time dimension in a star schema may be supported by a dimension table with the following structure: DAY_ID DAY_DESC MONTH_ID MONTH_DESC QUARTER_ID QUARTER_DESC YEAR_ID By contrast, a snowflake schema has a separate lookup table for each level of a dimension. In a fully normalized snowflake schema, each lookup table contains only its attribute's ID and description columns, and the ID of its parent to facilitate joins up the hierarchy. Lookup tables may also be partially denormalized, in which ID columns of all parents in the dimension are included, or fully denormalized, in which IDs and descriptions of all parents are included. Fully normalized snowflake schema: DAY_ID DAY_DESC MONTH_ID MONTH_ID MONTH_DESC QUARTER_ID QUARTER_ID QUARTER_...