Table of Contents

Publishing Your First Model

Once your model has tables, relationships, and measures, it's time to deploy it to a server so it can be used in Power BI reports and other tools.

Before You Publish

Make sure your environment's Publish Server is configured. This is the server where your finished model will live, separate from the Working Server you use during development.

  1. Click Environments in the activity bar.
  2. Select your environment.
  3. Under Publish Server, enter the server details for your target:
    • Microsoft Fabric: Your Fabric workspace URL
    • Azure Analysis Services: Your Azure AS server name (e.g., asazure://region.asazure.windows.net/servername)
    • SQL Server Analysis Services: Your SSAS instance name (e.g., servername\instancename)

Publishing

  1. Click Publish in the toolbar.
  2. Semantic Modeler deploys your model definition to the Publish Server configured in your active environment.
  3. Once published, the model is available for connections from Power BI Desktop, Excel, and other client tools.

Processing After Publishing

Publishing deploys the model definition (schema, relationships, measures), but the tables don't contain data yet. You need to process the model to load data:

  1. Click Process in the toolbar.
  2. Choose the processing mode:
    • Full: Drops and reloads all data. Use this for the first processing or when you want a clean refresh.
    • Default: Only refreshes partitions that are out of date.
    • Calculate: Recomputes calculated columns and measures without reloading source data.

For Fabric workspaces, Semantic Modeler shows per-table progress so you can see exactly which tables are done and which are still loading.

Fabric: Handling Missing Connections

When you publish to a Fabric workspace for the first time, the model's data source connections may not match the connections available in the workspace. If this happens:

  • Semantic Modeler attempts to automatically bind unbound data sources to matching connections in your workspace.
  • It prefers shared/organizational connections over personal cloud connections.
  • If automatic binding fails, you'll need to set up the connection in the Fabric portal first, then publish again.
Tip

Set up your data source connections in the Fabric workspace before publishing for the smoothest experience. This way, Semantic Modeler can bind to them automatically.

Publishing to SSAS

SQL Server Analysis Services works on all platforms, but connecting from macOS or Linux requires some extra setup because the underlying ADOMD.NET library uses Windows authentication.

If you're on macOS or Linux and need to publish to an on-premises SSAS instance, see Kerberos Authentication for SSAS for what's involved.

Note

Fabric and Azure Analysis Services use Azure AD token-based authentication, which works on all platforms without any extra configuration.

Verifying Your Deployment

After publishing and processing:

  1. Open Power BI Desktop (or Power BI in the browser for Fabric).
  2. Connect to your server/workspace.
  3. You should see your model with all its tables, columns, and measures.
  4. Build a quick report to confirm that data is flowing correctly and your measures return expected results.

What's Next

  • Environments: Set up additional environments for staging and production.
  • Source Control: Version control your model alongside your publishing workflow.
  • Fabric Sync: Keep your local model in sync with changes made by teammates in a Fabric workspace.
Edit this page