Question about moving to Serverless compute

[Inna_M]

Hi my organization is using Databricks in Canada Est region and Serverless isn't available in our region yet? at all?

I would like to know if it is worth the effort to change region for the Canada Central, where Serverless compute is available. 

We do not have big loads yet and use medallion architecture. We use Unity Catalog and planning to use Fabric, PowerBi for reporting

Some of our workflows use shared cluster and some use the job cluster, so it takes 4 minutes extra to start up. 

If the answer is Yes, what components should be moved to the other region (network, storage etc)?

Thank you

[Takuya-Omi]

@Inna_M 

Currently, serverless compute is not supported in CanadaEast.
Azure Databricks - Serverless Availability by Region

To use serverless compute, simply create your workspace and Unity Catalog metastore in a supported region.
Even if your storage is located in a different region, you can still access and manage data via external locations.

[Inna_M]

hi Takuya Omi, thank you for responding. My question is: if we are to migrate our existing workspaces (3) and UC to Canada Central. Is it doable? Is it worth it? What does it imply? What are the best practices to do so?

Thank you. 

[Rjdudley]

The best part about serverless is quickly attaching to compute.  You can kind of replicate this with pre-warmed pools: Connect to pools - Azure Databricks | Microsoft Learn.  The environments between serverless and classic compute are slightly different.

Migrating workspaces to a different region is a complete do-over, you have a fresh environment, and you have to create everything Databricks from the start (and maybe Azure depending on your company's policies).  That can be a lot of effort.  Databricks appears to be rolling out serverless region-by-region, so contact your account team and see if they know if/when your main region will get serverless.