Forum Discussion

BIXL's avatar
BIXL
Icon for Resolver I rankResolver I
8 years ago
Solved

Performance best practice with wide tables

Hey All, In general we know that wide tables=bad performance in Power BI models. Say for the sake of the example that I have a 500 columns table that contains a lot of measures and some attributes ...
Need Help
  • v-sihou-msft's avatar
    v-sihou-msft
    8 years ago

    BIXL

     

    Separating frequently used column and other useless columns so that it will not consume more memory when calling frequently used columns only. For the internal working of VertiPaq engine, you can refer to this article: The VertiPaq Engine in DAX 

     

    Regards,

     

     

BIXL's avatar
BIXL
Icon for Resolver I rankResolver I
8 years ago

Hi v-sihou-msft

Thanks for your reply.

I am aware of the best practices described bellow and in the link you provided.

My question was in regard to the way the VertiPaq engine works.

Since there are some unique situation where there are a lot of measures in a single table the question was if in engine performance point of view you should split them into several tables connected by the same PK. 

could you elaborate some more about :

"Depends on usage, you can also include frequently used columns into same table together" - Does this affect performance ? or just a more "user friendly" model, for which I'm totally on.

 

Thanks

v-sihou-msft's avatar
v-sihou-msft
Icon for Microsoft Employee rankMicrosoft Employee
8 years ago

BIXL

 

Separating frequently used column and other useless columns so that it will not consume more memory when calling frequently used columns only. For the internal working of VertiPaq engine, you can refer to this article: The VertiPaq Engine in DAX 

 

Regards,