Forum Discussion

pennyhoho117's avatar
pennyhoho117
Icon for Helper IV rankHelper IV
1 year ago
Solved

User wants to use two different Semantic Model Data to build report, how to do that

We have two different semantic models, and user wants to use the two semantic model data to build a report.  Can we use two  diff. model in same report? if Yes, how to do it?
  • v-veshwara-msft's avatar
    1 year ago

    Hi pennyhoho117 ,
    Thanks for reaching out in Microsoft Fabric Community.
    Thanks Akash_Varuna for your inputs.

    To build a report using two different semantic models, you can follow one of the methods below:
    Method 1: Composite Models with DirectQuery in Power BI Desktop 

    1. Open Power BI Desktop.
    2. Click on Get Data -> Select Power BI semantic models.
    3. Choose the first semantic model and click Connect.
    4. Repeat the process to connect to the second semantic model.
    5. Power BI will force DirectQuery mode for both datasets.
    6. Navigate to Model View and create relationships if necessary.
    7. Build your report using data from both semantic models.
    8. Publish the report to Power BI Service.

    Note:
    When combining multiple Power BI semantic models, both use DirectQuery, and Import Mode is not allowed. If you need to mix Import Mode and DirectQuery, you should connect to a database like SQL, Lakehouse, or Warehouse instead of another semantic model.
    DirectQuery can impact performance since queries are executed directly against the source models.

    Method 2: Using Dataflows in Power BI Service

    1. Identify the original data sources (Lakehouse, SQL Database, Warehouse) for each semantic model.
    2. In Power BI Service, create a new Dataflow, then click on Get Data and select the appropriate Lakehouse, Warehouse, or Database where the semantic model is associated.
    3. Choose the authentication method and click Connect.
    4. Repeat the process for the second semantic model.
    5. Combine Data – Use the Combine option in the top ribbon to:

      • Append Queries (to stack data from both sources).
      • Merge Queries (to join data based on a common key).
    6. Add a data destination (Lakehouse, Warehouse, SQL Database, or Kusto Database) and publish the Dataflow.
    7. Go to the SQL endpoint of the destination database to build the report, or use Power BI Desktop to connect to the published Dataflow, build the report, and publish it to Power BI Service.

    Hope this helps. Please reach out for further assistance.

    If this post helps, then please consider Accept it as the solution to help the other members find it more quickly and a kudos would be appreciated.

    Best Regards,
    Vinay.