Calculated Table causes performance issue and dataset refresh fails

[sivarajan21]

Hi,

I have a calculated table called 'Invoice Issues' as below which is derived from the existing Data table:

The main purpose of this table is to create a calculated column called 'Status' which is used in visual:

Calculated column:

visual:

This works perfectly and our report is liked by the customers.But this causes lot of performance issues and dataset report refresh fails in workspace.

Is there a way to creat this as a dax measure or any other alternate solutions is much appreciated?

PFA file herewith PR-419 - Data Coverage RLS ADB (1).pbix

@amitchandak @Ahmedx @marcorusso @Greg_Deckler @Ashish_Mathur @Anonymous 

[Ashish_Mathur]

Hi,

Share a much smaller sample dataset with only the relevant tables.  Explain your requirement and show the expected result.

[marcorusso · Accepted solution]

When you create a calculated table, the entire uncompressed table must be materialized in memory and then compressed. This could be a memory-intensive operation that works on your PC and fails on Power BI Service.

You should reduce that table by removing unused columns and remove unnecessary rows - e.g. keep only the rows that have an issue.

Or compute the column outside of Power BI, so it's processed like other tables segment by segment, reducing the memory requirement at refresh time.

[sivarajan21]

Hi @marcorusso 

Many thanks Sir for your quick response!

Apologise for delay in response! we will follow this going forward in our reports

Just wanna confirm,
 when you say outside of power bi, it means in our case it is Data Flow where transformations can be done?

when you say 'keep only the rows that have an issue.' that means applying Date range parameters in power query?

Is there a way to measure the performance of data model & dataflow similar to how dax studio is used for measuring the power bi report performance?

 Thanks in advance!

@amitchandak @Ahmedx @marcorusso @Greg_Deckler @Ashish_Mathur @v-cgao-msft @Daniel29195