Delete a Data Source in Solution Explorer (SSAS Multidimensional)

Applies to: SQL Server Analysis Services Azure Analysis Services Fabric/Power BI Premium

You can delete a data source object to permanently remove it from an Analysis Services multidimensional model project.

In SQL Server Analysis Services, data sources provide the basis on which data source views are constructed, and data source views are in turn used to define dimensions, cubes, and mining structures in an SQL Server Analysis Services project or database. Therefore, deleting a data source may invalidate other SQL Server Analysis Services objects in an SQL Server Analysis Services project. You should always review the list of dependent objects that is provided before deleting the object.

Important

Data sources on which other objects depend cannot be deleted from an SQL Server Analysis Services database opened by SQL Server Data Tools in online mode. You must delete all objects in the SQL Server Analysis Services database that depend on that data source before deleting the data source. For more information about online mode, see Connect in Online Mode to an Analysis Services Database.

To delete a data source

  1. In SQL Server Data Tools, open the project or connect to the database from which you want to delete a data source.

  2. In Solution Explorer, expand the Data Sources folder.

  3. Right-click the data source, and then click Delete. The Delete Objects dialog box appears, showing the objects that will be invalidated if you delete the data source. Review this list carefully before clicking OK to delete the data source.

  4. Save the project.

    After deleting a data source from an SQL Server Analysis Services project, you must save the modified project or you will receive an error the next time you open the project because the underlying XML file for the deleted data source will be missing when the project attempts to load the deleted data source.

See Also

Data Sources in Multidimensional Models
Supported Data Sources (SSAS - Multidimensional)