Forum Discussion
SQL Server Stored Procedure and SSIS Package migration to Microsoft Fabric
We have a requirement to migrate on-prem EDW - sql server stored procedures & SSIS packages to Microsoft Fabric. What would be the best solution approach for this?
Microsoft Fabric does not provide an SSIS connector as of today, so would it be convenient to migrate the SSIS packages to Azure Data Factory and then from Azure Data Factory to Fabric Data Factory ?
Also what would be best solution approach to migrate the on-prem stored procs to Fabric Warehouse stored procs? Regards!
Tx......Avishek
4 Replies
Hi,
About the SSIS, I think you will have no choice but rewritting the packages as pipelines. Even if you migrate to Azure Data Factory, they will not integrate with Fabric.
In relation to the stored procedures, did you tried to script them and check if all the code on them is compatible with Fabric Data Warehouse ?
Kind Regards,
Dennes
Stored proc activity is already there in Fabric.
Run SSIS package, currently, only available in ADF.
For Fabric, you can vote on Fabric Idea: SSIS integration runtimes in Fabric Data Factory for the need. We are working on the experience to bring SSIS workloads in Fabric Data Factory. And will keep update on idea when we have concrete date to share.
Native tooling for large-scale SSRS / SSIS → Fabric migration is still quite limited, which is why many teams struggle with manual recreation and validation.
We’ve addressed this using a Gen AI–powered Fabric Migration Copilot that automatically convert SSIS pipelines, ingestion logic, and reporting artifacts into Fabric-native components, while also validating visual and data fidelity.
This has helped customers cut migration effort by ~60% and accelerate timelines by ~50%, while minimizing regression risks.
Want to know more about the solution ? Write to me at [email protected]if you have budget consider carefully this process because it could be a longer journey then expected.
you can't do lift and shift and you're jumping on a different platform.
for example, if your datawarehouse runs on SQL Server, in Fabric you'll find Warehouse that behind the scenes runs SQL Server in a distributed environment therefore the same T-SQL may run faster, slower or not run at all (there's no 100% compatibility for T-SQL).
moreover you must keep in mind there're unsupported features (enforced constraints, indexes, etc.).
you may also consider to revist patterns, things you did in multiple steps in SSIS may become one single activity in Data Pipeline or would better to approach it in a code way with T-SQL (copy into) or notebooks to lower CUs usage.
and much more...
