Building efficient data models

We will begin with some theoretical concepts on how to model data for fast query performance. These techniques were designed with usability in mind but happen to be the perfect way to model data for the Analysis Services engine in Power BI. We will begin by introducing star schemas because they are native to the Analysis Services engine and it is optimized to work with them.

The Kimball theory and implementing star schemas

Data modeling can be thought of as how to group and connect the attributes in a set of data. There are competing schools of thought as to what style of data modeling is the best and they are not always mutually exclusive. Learning about competing data modeling techniques is beyond the scope of this book.

We will be looking at dimensional modeling, a very popular technique that was established by the Kimball Group over 30 years ago. It is considered by many to be an excellent way to present data to business users and happens...