Performance REMOVEFILTERS and SUMX

[Anonymous]

I am concerned performance-wise for the next change in a measure:

Change... From:

Revenue LYTD:= CALCULATE([Revenue YTD] , SAMEPERIODLASTYEAR('Date'[Date]) )

To:

Revenue LYTD:=
VAR VAR1 = 
ADDCOLUMNS(VALUES(Revenue[Key_Customer]), 
                 "Col",CALCULATE([Revenue YTD],SAMEPERIODLASTYEAR('Date'[Date]),REMOVEFILTERS(Revenue[Segment],Revenue[Segment Type])))

RETURN SUMX(VAR1, [Col])

I am testing with little data, and it works fine; but is there any performance-penalty for forcing filter removal with REMOVEFILTERS?

[VahidDM]

Hi @Anonymous 

I don't have any idea about your data model and report, but I think the below measure value would be the same. Would you please test this

Revenue LYTD :=
VAR VAR1 =
    SUMMARIZE (
        Revenue,
        Revenue[Key_Customer],
        "Col", CALCULATE ( [Revenue YTD], SAMEPERIODLASTYEAR ( 'Date'[Date] ) )
    )
RETURN
    SUMX ( VAR1, [Col] )

then you can measure the performance :

https://docs.microsoft.com/en-us/power-bi/create-reports/desktop-performance-analyzer#:~:text=In%20Power%20BI%20Desktop%20select,display%20the%20Performance%20Analyzer%20pane.

If this post helps, please consider accepting it as the solution to help the other members find it more quickly.

Appreciate your Kudos!!