thanks for sharing those different options — very helpful to see multiple angles explored.

I’m also looking into this for Microsoft Fabric Warehouse, and from what I can tell, a lot of the traditional approaches (like using. sys.objects, sys.partitions, or even Don’tIt behaves quite the same way in Fabric compared to classic Azure SQL Data Warehouse or SQL Server.

The dynamic SQL route using sp_executesql It is promising, but as SRP mentioned, performance becomes a bottleneck with thousands of tables. Ideally, Fabric should expose a system view similar to Butt so far, it looks like that's not available.

If anyone has insights into whether there's an equivalent metadata view in Fabric (like a DMV or any internal table that tracks row counts), that could be a cleaner way forward.

It would be great to hear from anyone who has tested this specifically on Fabric.