Matrix Performance

[Anonymous]

Hey, so i'm working with a dataset that has over 440 million rows. I cant import it so I've been using Direct Query. The issue I'm running into is that my managers want around 15-20 measures (total sales, YoY growth, YTD, total unit % change, etc.) and the matrix takes a really long time to load when i have more than 5 measures. I tried creating variables within measurse but this didnt improve speed. my next thought is to create a drillthrough and just display less information but i'm wondering if there is a way for me to display 10-15 measures in a matrix while not sacrificing performance?

Thanks in advance!

[V-pazhen-msft · Accepted solution]

@Anonymous 

There is no way to maintain the performance if you use measure, my suggest would be to convert some measures into columns, so the power bi pre-calculates the result when you open the pbix instead of calculates when creating matrix visuals.

Paul Zheng _ Community Support Team
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
 

[Greg_Deckler]

@Anonymous - I would recommend looking into Aggregation Tables. https://docs.microsoft.com/en-us/power-bi/transform-model/desktop-aggregations#:~:text=The%20aggregation%20feature%20interacts%20with%20table-level%20storage%20modes.,non-multidimensional%20DirectQuery%20data%20sources%20can%20work%20with%20aggregations.