Deployement Pipeline support for Dataflow Gen2

[chandulae]

Hi,

We are experiencing significant issues with deploying Dataflow Gen 2 from development to production.

Currently, we are using the manual export and import method, saving them to a local drive.

Handling over 100 tables makes this manual method very time-consuming.

Could we have support for a deployment pipeline with Dataflow Gen 2? If it is already planned, could we know the release date?

Thank you.

[miguel]

@TBNPieter we're currently working on some new features aimed towards addressing those scenarios that you've described, but they're not based on deployment rules.

Instead, these other features are more intrinsic to how the Dataflow will just work by default leveraging relative references so the need for a deployment rule won't be needed. We're also working on enabling support for workspace variables which also enables multiple other scenarios where a deployment rule would be unnecessary. 

I'm hoping that we can share mroe in the near future, but just want to confirm that we plan to address those scenarios in a much cleaner way that is quite integrated with the rest of Fabric so that you can pick either relative references (intrinsic to the Mashup document in your Dataflow) OR workspace variables which are more global and can control multiple Fabric items at the same time.

[Cordeniz]

The lack of this feature is a big problem when using the deployment pipelines. Each time I transport developments between workspaces, I will need to change the target of each table manually.

[miguel]

This is currently available. We also have a documentation article that can help you understand the available solution architectures that support CI/CD scenarios that you could leverage through deployment pipelines.

CI/CD and ALM solution architectures for Dataflow Gen2 - Microsoft Fabric | Microsoft Learn