Introduction

So far, we've been through the basics of MDX calculations. We learned a few tricks regarding time calculations, practiced making concise reports, navigated hierarchies, and analyzed data by applying typical business calculations. Now, we will discuss special topics.

This is a book that follows the cookbook approach, the main topic being MDX. In this chapter, however, we're going to discover that MDX calculations are not always the best solution. It is only one of the possible layers we can start from. The other two layers are the cube design and the underlying data warehouse (DW) model. Each layer depends on the other. A good data model will enable a good cube design which in turn will enable simple MDX calculations.

Whenever we are given a request to get something from the cube, it is not only a request to make an adequate MDX query or calculation and then to return the data, it is something much deeper-a challenge to our cube design and the underlying dimensional model. If the...