Schema changes in delta table impacting semantic model and PowerBI reports

[prantika]

We have a use case which comprises renaming fields in the existing delta tables. These tables are ingested through Direct Lake connection in semantic model and utilized for creating PowerBI reports in Fabric.

We tried renaming one of the table, which caused to break existing relationships with the tables and add the renamed table as a new table. This is affecting our PowerBI reports that uses DAX measures.

What is the optimized method to perform schema changes in tables, which does not break the DAX measures and existing relationships created in the semantic model? Is there any way to replace tables in semantic model that allows us to map field between these tables post schema changes?

Thanks!

[Anonymous · Accepted solution]

Hi @prantika ,

According to your description, it seems that the table was renamed of delta tables. When making schema changes to delta tables that are used in a semantic model and PowerBI reports, it's important to consider the impact on existing relationships and DAX measures. Please update the following info in Power BI:

• Update the related table info in semantic model
• Check and recreate the relationship if there is any relationship created base on the field of the renamed table
• Update the measure if there is any measure reference the renamed table

And you can refer the following link for optimizing schema changes in Delta tables:

Delta Lake table optimization and V-Order - Microsoft Fabric | Microsoft Learn

Best Regards