Forum Discussion

Bilge's avatar
Bilge
New Member
8 months ago
Solved

Power Bi Report Server

I have a 50GB SSAS Tabular Model. Is it logical to create it in a Power BI report or to connect to the tabular model with a direct query? I am using Power BI Desktop RS May 2025 version.
powerbi
SSAS
SSIS
  • V-yubandi-msft's avatar
    V-yubandi-msft
    8 months ago

    Hi Bilge ,

    When working with a 50GB SSAS Tabular model and a large user base, direct querying can cause performance challenges. Here are several strategies to help address this.

    1. Pre aggregate frequent queries in SSAS so Power BI accesses smaller, faster datasets.

    2. Divide the model into partitions, such as by date or region, to focus queries and improve speed.

    3. Refresh only updated data instead of the entire model.

    4.  Use Import mode for smaller, commonly used data and DirectQuery for larger tables.

    5.  For SSAS Enterprise users, enable scale-out to balance query loads across multiple servers.

    Refine DAX measures, minimize calculated columns, and ensure efficient relationships to reduce query time.

     

    Applying these approaches can help maintain your SSAS model's integrity while keeping Power BI performance high for users.

     

    Regards,
    Yugandhar.

Zanqueta's avatar
Zanqueta
Super User
8 months ago

Hi Bilge,

 

Thinking about your question, in my oppinion:


A tabular model of 50 GB is extremely heavy for Power BI Report Server. This results in:
  • High memory consumption: The VertiPaq engine must load the entire dataset into RAM, which can saturate the server.
  • Very slow refresh times: Each refresh could take hours and affect other services.
  • Limited scalability: Any further growth makes the model even harder to manage.
For these reasons, importing this volume into Power BI maybe could be not the best option
 
Instead of attempting to load everything into Power BI, consider revising the architecture and adopting a more modern and scalable solution:
 

Use Microsoft Fabric

  • Create a Lakehouse or Data Warehouse in Fabric to store the data in an optimised structure.

Alternative with SQL Server

  • Expose the data through optimised views or stored procedures in SQL Server.
  • Use DirectQuery in Power BI to avoid loading 50 GB into the model.
 
Official References:

DirectQuery in Power BI: When to Use, Limitations, Alternatives - Power BI | Microsoft Learn

 

If this response was helpful in any way, Iā€™d gladly accept a šŸ‘much like the joy of seeing a DAX measure work first time without needing another FILTER.

Please mark it as the correct solution. It helps other community members find their way faster (and saves them from another endless loop šŸŒ€.