MicrosoftI have used the Meta-Data Driven Copy Task in Azure Data Factory, and for that we typically set up a Control Table connection to a dedicated Azure SQL Database if the architecture does not also include an Azure SQL Database or Synapse Warehouse.
But with the advent of Fabric, many clients are trying to get away from needing all these stand-alone resources (read: Azure SQL DB, Data Warehouses, Pipelines, etc.). After all, that is what Microsoft is trying to sell us with Fabric: EVERYTHING under ONE ROOF, easily accessible and seamlessly connected. But there is a whole host of limitations in Fabric that preclude us from actually using it to replace the Meta Data Driven Copy Task in classic ADF (like no NVARCHAR(MAX) data type, no support for MERGE operation, and the list goes on).
I, too, can dream of a day where Microsoft publishes a wizard in Fabric that sets up the Top/Middle/Bottom level pipeline structure, all in Fabric. Oh, and while you're at it, Microsoft, please re-factor the thing so it supports TIMESTAMP datatypes as the Watermark. (While the wizard does not support it, we have been able to successfully re-vamp the bottom level pipeline so that it can use a TIMESTAMP as the Watermark.)
