Before I start writing this article, I want to simply suggest to use star schema. However, someone may not build a dimensional data warehouse, so I generalize my words to denormalization. Per performance concern, we always use denormalized data model to support fast user queries.
In relational data warehouse, the data model is normalized. No worry, it is perfect per design. Just make sure no user granted to access this layer since it doesn't support good user queries performance. We should build denormalized data marts on top for user queries.
In dimensional data warehouse, simply use star schema instead of snowflake if possible. And try our best to avoid those worst practice in dimensional modelling such as "Fact table joining fact table" nor huge size dimension.
No comments:
Post a Comment