DAX Calculate Efficiency of Filter on Fact vs. Related Table

[cr731]

I have a fact file containing a column of measures and corresponding values, for example,

[TABLE="class: grid, width: 500"]
<tbody>[TR]
[TD]Measure[/TD]
[TD]Value[/TD]
[/TR]
[TR]
[TD]Volume[/TD]
[TD]100[/TD]
[/TR]
[TR]
[TD]Revenue[/TD]
[TD]70[/TD]
[/TR]
[TR]
[TD]Cost[/TD]
[TD]20[/TD]
[/TR]
</tbody>[/TABLE]

This is repeated for thousands of transactions across many months - so my table is a few hundred thousand rows.

I want to write measures for each of the "Measures" in my fact file, for instance

Sum of Volume
Sum of Revenue
Sum of Cost

Is it more efficient to do this

= CALCULATE( SUM( Fact[Value] ), Fact[Measure] = "Volume" )

or better to create a related table of the list of unique Measures

Volume
Revenue
Cost

And then write my calculate formula as

= CALCULATE ( SUM Fact[Value], RelatedMeasuresTable[Measure] = "Volume" )

Is one more efficient than the other?

Thanks

 

[gazpage]

I'm struggling to see the benefit of the latter approach. Why would that be quicker?

A couple of hundred thousand rows should be no problem anyway.

 

[cr731]

I'm struggling to see the benefit of the latter approach. Why would that be quicker?

A couple of hundred thousand rows should be no problem anyway.

I just wasn't sure - I thought I read somewhere that filtering on a related table containing just the unique values was supposed to be better than filtering on the fact table containing many duplicates in a given column. Maybe it doesn't matter much in this case. Thanks