Introduction
Once the framework is set up, it's time to focus on the different layers of our data warehouse. There are various architectural schools of thought when it comes to data warehouses:
- Corporate Information Factory (CIF)
- The Kimball Group dimensional data warehouse
- Data vault
The main difference between the Kimball Group and the others is the way a datamart is loaded. The Kimball Group approach loads data into a staging area and from there, refreshes the data warehouse. The latter is modeled as a dimensional data warehouse. It is also known as a datamart or star schema. The Kimball Group approach uses denormalized tables in its data warehouse.
A typical data warehouse using the Kimball Group method has the following components:
- Data sources that can be in different formats such as text files, databases, Excel, and so on
- A staging area that can be either persistent (contains all history of data loaded) or transient (emptied every time data is loaded)
- One or more datamarts that are tied to...
