Azure Analysis Service - Memory consumption per Model

Alex H 31 Reputation points
2021-09-01T10:00:17.407+00:00

We are runing on Tier S2 with a limit of 50GB, but once in a while we exceeded this limitation and so we upgraded temporary to S3 for a few days and then downgraded back again to S2.
After this switch from S2 to S3 and back again to S2, both times the memory consumtion collapsed and sticked to a size below the S2 limitation.

128196-image.png

Now I'm trying to figure out what causes this high amount of memory consumtion. But I'm only see the memory of the whole AAS.
I'm not able to add a split for each Model.

128253-image.png

As you can see, we have 17 models:

128254-image.png

So my questens are :
1.) How can I split the view, so that I can see the memory consomption for each model
2.) How can I figure out which memory is used by the models and which is only used by the system (temporary, working memory, etc.)

Azure Analysis Services
Azure Analysis Services
An Azure service that provides an enterprise-grade analytics engine.
439 questions
Microsoft Entra ID
Microsoft Entra ID
A Microsoft Entra identity service that provides identity management and access control capabilities. Replaces Azure Active Directory.
19,655 questions
Accepted answer
  1. HimanshuSinha-msft 19,381 Reputation points Microsoft Employee
    2021-09-14T15:54:44.983+00:00

    Hello @Alex H ,
    Thanks for the patience , unfortunately we are not getting the kind of response from the team here . if you have a support plan you may file a support ticket, else could you please send an email to azcommunity@microsoft.com with the below details, so that we can create a one-time-free support ticket for you to work closely on this matter.
    Subscription ID:
    Subject : Attn Himanshu
    Please let me know once you have done the same.

    Update Oct 21

    I was reviewing the request and yiou did worked with some one from support . I am sharing the resolution on a very high level so that it can be useful to other team members .
    After investigation it was found that the reason behind the high memory utilization on the server is heavy load when compared to the SKU limit.

    Thanks
    Himanshu

1 additional answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. Nandan Hegde 29,896 Reputation points MVP
    2021-11-09T06:20:53.41+00:00

    Hey @Alex H ,
    An another way can be to set up diagnostic settings on AAS and map it to log analytics.
    https://learn.microsoft.com/en-us/azure/analysis-services/analysis-services-logging

    And via log analytics query, you can get that details split 

    search * | where ( Type == "AzureDiagnostics" ) | where ( EventClass_s == "QUERY_END" ) | project Category,ResourceGroup,Resource,ResourceType,OperationName,DatabaseName_s,Duration_s,TextData_s,User_s,CPUTime_s

    @HimanshuSinha-msft : Any thoughts on this?

    0 comments