Forum Discussion

FabBI_123's avatar
FabBI_123
Frequent Visitor
1 year ago

Performance in Analyse in Excel

Hi, 

One of our old SSAS cube (Multidimentional) cube has been converted to fabric semantic model currently using F64

users need to use their old expeirence using Excel endpoint because of that we have used analysis in Excel feature in Fabric Semantic model.

 

When we compared performance between two, SSAS took less time than Fabric analysis in Excel for 'OLAP running'  in granular levels. (with PowerBI there is no issue).

 

Im aware that multidimention calculate the values in the processing period and tabular using on the fly calculations.

 

Does there a performance lack in Excel endpoint when comparing Multidimention and Tabular models generally ?

 

Is there a way to overcome it ?

i did find the below article  : Chris Webb's BI Blog: Excel PivotTable Query Performance With Analysis Services, Power Pivot And Power BI (Revisited) (crossjoin.co.uk)

 

 

 

1 Reply

  • Power BI has to emulate MDX when running Analyze in Excel. That translation impacts performance.

     

    Nowadays our power users have moved away from Analyze in Excel towards "Live connection"  which is based on DAX queries against the Semantic Model.  Even though they lose the cube access they seem to be happier.