How to Diagnose a Slow Power BI Report by Jason Romans

Users complain that a Power BI report is so slow that it is not worth using. A Power BI report may not perform well for many reasons, but how do you know where to start? In this session, I will show you some of the main tools I use to dive deep into the common and the not-so-common performance issues I run into. Starting from a slow report, we will learn how to diagnose where the problem is. Beginning with Performance Analyzer in Power BI, we can start to figure out where the issue is. Is it the visual that is slow, or are the DAX queries taking too long?

Next, we will use DAX Studio to tune the DAX queries that are being generated. Along the way, we will look beneath the visuals to see how Power BI is architected so we can better understand how to write better code from the start. We will examine Query Plans and Server Timings to evaluate which of our changes are more beneficial.

Lastly, we will examine methods for detecting and preventing performance issues before they occur. Using Tabular Editor, we will learn how Best Practice Analyzer can detect performance issues with your semantic models.