Memory and performance considerations
The design of a Power BI model highly impacts its size, and size is highly correlated with performance. In this section, we share some best practices to optimize the performance of your model, as a recapitulation of the topics discussed in this chapter. As a rule of thumb, smaller models with respect to size are faster. You can use the file size of the Power BI model as an indication; you can also get a more detailed view of size and performance by using specific community-driven tools like DAX Studio.
Keep in mind the following guidelines while designing your Power BI model:
- Having fewer columns is better. The Power BI model achieves a high compression rate of data due to the columnar database concept. However, it still needs to keep track of which values belong together in a row. The more columns a table has, the more overhead the model needs to know what goes where. So, keep the number of columns per table as small as possible...