02-dimension-patterns

Handle NULL values in dimension tables by using dedicated "Unknown" member rows. This enforces best practices for NULL handling in dimensional data warehousing, ensuring referential integrity and preventing BI reporting errors.

Choose between junk, mini, and outrigger dimensions by evaluating your specific modeling needs. Use junk dimensions for low-cardinality flags, mini dimensions for rapidly changing attributes, and outrigger dimensions to normalize dimension hierarchies.

A role-playing dimension is a single dimension table referenced multiple times in a fact table using different foreign keys. Use it in data warehousing when the same dimension context applies to multiple business events, such as order date and ship date.

Flatten brand hierarchies directly for simplified BI querying, or use an outrigger dimension to normalize data and reduce storage. The choice depends on whether you prioritize query performance or data normalization in your dimensional model.

Manage low-cardinality flags in a dimensional model by storing them as textual attributes. This approach improves BI readability and allows them to be grouped efficiently into junk dimensions for better gold layer organization.

A degenerate dimension is a dimension key that has no corresponding dimension table, typically stored as a transaction number in the fact table. It fits dimensional modeling by preserving control numbers without creating redundant gold layer tables.