The underlying data model used for DAX is the in-memory tabular cube. Since this data model is designed to be used in memory, the overall size of the data model is an important performance consideration since, after all, computers only have a finite amount of memory. Understanding the underlying workings of the tabular data model is imperative to building efficient data models that fit into available memory and optimize the speed and performance of DAX calculations. This recipe provides guidance on how to optimize your data model for peak efficiency.