Chapter 10: Data Modeling and Row-Level Security
In the previous chapter, we looked at the visual layer in Power BI, where a key point was to reduce the load on data sources by minimizing the complexity and number of queries. We learned that this area is usually the easiest and quickest place to apply performance-related fixes. However, experience working with a wide range of Power BI solutions has shown that issues with the underlying dataset are very common and typically have a greater negative performance impact. Importantly, this impact can be amplified because a dataset can be used by more than one report. Dataset reuse is a recommended practice to reduce data duplication and development effort.
Therefore, in this chapter, we will move one layer deeper, into modeling Power BI datasets with a focus on Import mode. Dataset design is arguably the most critical piece, being at the core of a Power BI solution and heavily influencing usability and performance. Power BI's feature...
