February 2019
Beginner to intermediate
442 pages
9h 38m
English
The dimensional modeling technique uses facts and dimensions to build the data model. This modeling technique was developed by Ralf Kimball. Unlike ER modeling, which uses normalization to build the model, this technique uses the denormalization of data to build the model.
Facts, in this context, are tables that store the most granular transactional details. They mainly store the performance measurement metrics, which are the outcome of the business process. Fact tables are huge in size, because they store the transactional records.
For example, let's say that sales data is captured at a retail store. The fact table for such data would look like the following:
A fact table has the following characteristics: