Forum Discussion
Fabric Deployment Pipeline Query
In a Fabric Deployment Pipeline is it possible to have workspaces that reside in different Fabric Capacities all within your own tenant?
We are currently using Fabric items within a Power BI Premium Capacity but looking at migrating to Fabric capacity in the next couple of months.
We are thinking of having an F64 (equivalent to our Premium at the minute) and a smaller pay as you go capacity, possibly an F16 for development purposes.
Will we be able to deploy across the capacities? Any thoughts on this and indeed the proposed architecture would be welcome.
Hello StitoTheGreat,
You currently:
Use Power BI Premium capacity (P SKU) with existing Deployment Pipelines.
Plan to migrate to Microsoft Fabric capacities (F SKUs), possibly:
F16 → for Dev/Test (pay as you go)
F64 → for Production (reserved instance)
Your goal: use Deployment Pipelines across these workspaces, which would reside in different Fabric capacities — still within the same tenant.
Yes, it is possible to include workspaces hosted on different Fabric capacities in the same Deployment Pipeline, as long as all capacities belong to the same tenant.
Microsoft explicitly supports cross-capacity deployment within a tenant under these conditions:
All capacities (e.g., F16, F64) are under the same tenant and region.
All workspaces are Fabric-enabled (i.e., not legacy shared capacity).
You have Contributor/Admin permissions on all target workspaces.
Source :
- Fabric Deployment Pipelines – Workspace requirements
- https://learn.microsoft.com/en-us/fabric/cicd/cicd-overview
- https://learn.microsoft.com/en-us/fabric/admin/capacity-settings?tabs=power-bi-premium
Hope it can help you !
Best regards,
Antoine
Exactly, it’s considered best practice to host your different environments (Dev, Test, Prod) on separate Fabric capacities.
That way, any heavy or unstable development workloads can’t impact your production performance or consume shared compute.You can monitor and tune usage on your Dev capacity using the Fabric Capacity Metrics App, and once everything is validated and optimized, you can safely promote and deploy the content to the Production capacity through your deployment pipeline.
3 Replies
- AntoineW
Super User
Hello StitoTheGreat,
You currently:
Use Power BI Premium capacity (P SKU) with existing Deployment Pipelines.
Plan to migrate to Microsoft Fabric capacities (F SKUs), possibly:
F16 → for Dev/Test (pay as you go)
F64 → for Production (reserved instance)
Your goal: use Deployment Pipelines across these workspaces, which would reside in different Fabric capacities — still within the same tenant.
Yes, it is possible to include workspaces hosted on different Fabric capacities in the same Deployment Pipeline, as long as all capacities belong to the same tenant.
Microsoft explicitly supports cross-capacity deployment within a tenant under these conditions:
All capacities (e.g., F16, F64) are under the same tenant and region.
All workspaces are Fabric-enabled (i.e., not legacy shared capacity).
You have Contributor/Admin permissions on all target workspaces.
Source :
- Fabric Deployment Pipelines – Workspace requirements
- https://learn.microsoft.com/en-us/fabric/cicd/cicd-overview
- https://learn.microsoft.com/en-us/fabric/admin/capacity-settings?tabs=power-bi-premium
Hope it can help you !
Best regards,
Antoine
Is it best/good practice to split the environments across capacities in this way e.g. separate capacity for DEV, Production etc.
I'm thinking it's a good way to protect the production resource from something that goes wrong in a development workspace that could potentially eat it up (we hve surge protection etc enabled).
How have people out in the community configured this? tia..
- AntoineW
Exactly, it’s considered best practice to host your different environments (Dev, Test, Prod) on separate Fabric capacities.
That way, any heavy or unstable development workloads can’t impact your production performance or consume shared compute.You can monitor and tune usage on your Dev capacity using the Fabric Capacity Metrics App, and once everything is validated and optimized, you can safely promote and deploy the content to the Production capacity through your deployment pipeline.
