In the previous tutorial, we learnt about facts and dimensions. In this tutorial, we would learn how fact and dimension tables are used together.
Star Schema
In dimension modelling, both the fact table and dimension table are grouped together. The fact table containing numerical measurements can be joined to a dimension table having descriptive texts. When the fact is joined with dimensions, it appears as a shape of star and hence is called Star Schema or Star Join Schema.
This schema is easily understandable and recognizable to business users. It is efficient as it has fewer joins.
Snowflaking
As discussed briefly in data warehouse tutorial-4, snowflaking requires creating separate tables. In simple words, one can say that there is further categorization of dimension table. Shown below is a snowflaked product dimension:
For example, product dimension is expanded further on the basis of brand, category and department. The further sub divided dimension has a primary key through which it is joined to the previous dimension table. In the previous dimension table, this key would thus be referred to as foreign key.
For example, brand key in product dimension is the foreign key of brand dimension. Similarly, category key in brand dimension is the foreign key of category dimension.
Continue reading the next tutorial for studying the basic components and data flow in data warehouse.
