Performance issues in powerBI

[f_r_1998]

 

Hello everyone,

I created a dashboard with 4 pages based on a semantic model. The data model is basically just one table with multiple columns. I created a couple of more independent tables for the measures and then also to do switchs. None of the tables are related to each other. In each page I am using a map, a table, some slicers, KPI cards and bookmarks. It is really not that much, so it seems weird to me that the performance is that bad. I used the performance analyzer and for some visuals it take up to 10000ms to load. Most of it is used for "other". I read online this refers to the interaction between the visualizations, so I deactivated all the interactions that were not necesarry. I also combined a lot of measures to have less in total and tried to switch some measure into calculated columns but I got system errors saying that there is an issue in processing. Despite these changes, the performance is still pretty bad. Does anyone have an idea what else I could try?

I also tried to replace the map with the ArcGIS Map for Powerbi and is working a bit better. However, I need to exclude the two options on the upper left (see screenshot). Is there any way to do this?

[Kedar_Pande]

It sounds like you've done a lot already!

• Trim your data to keep only necessary columns and rows.
• Use aggregated tables to create summaries that speed things up.
• Simplify your measures; less complexity can mean faster loading.
• Limit the number of visuals on each page to help with load times.
• Check your data types and use the most efficient ones for your columns.

Give these a shot and see if they help!