Forum Discussion

Anonymous's avatar
Anonymous
Not applicable
4 years ago

Report Perfomance slow cause by SSAS partitions ?

Hello All,

Is that report slowly cause by I used SSAS partitions ?
And also, I do a lot of testing in SSAS partitions like below table shows 

CPU

Core

Partition

Number

Report

time

(AVG)

Granularity
818 sec3 years
202008~202208
8220 sec202008~2021
202201~202208 
8318 sec 202008~202012
202101~202112
202201~202208


There is some detail in my test
1. Order topic have order header and order body two Fact table
2. there 6 million in header, 10 million in body

3. partition only do in header table 

4. I used 2019 SSAS 1500 level project 

5. I used DAX Studio to get report time

In MS doc partition limits said: Query performance can be negatively affected when the storage engine has to scan an excessive number of segments. But I only split 2 or 3 segment.

and can anyone explain why when I used partition feature then the report speed slowly 2 times? Thanks!

2 Replies

  • v-henryk-mstf's avatar
    v-henryk-mstf
    Icon for Community Support rankCommunity Support

    Hi Anonymous ,

     

    For storage configuration, you can specify the storage mode and whether to use active caching to control the latency of partitions.


    Refer to the related documentation below.

    Partition Storage Modes and Processing | Microsoft Docs

    Proactive Caching (Partitions) | Microsoft Docs

     

    If the problem is still not resolved, please provide detailed error information and let me know immediately. Looking forward to your reply.


    Best Regards,
    Henry


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

     

    • Anonymous's avatar
      Anonymous
      Not applicable

      Hi v-henryk-mstf ,

      Thanks for your reply, but I used Tabular Model not Multidimentional Model.

      Tabular Model is storage in memory and Multidimentional Model is storage in disk, in your refer to the related documentation their talk about Multidimentional Model