Databricks SQL warehouse best practices

[Phani1]

How best we can design Databricks SQL warehouse for multiple environments, and multiple data marts, is there any best practices or guidelines?

[Vinay_M_R]

• you can create at least 1 warehouse per business unit for cost controlling 
• you may want to have multiple WHs within the same BU in order to support different SLAs (i.e. 1 WH for interactive analyst/developers and 1 to support a production Dashboard)
• you don't need to separate WHs for data governance purposes, since data access is managed by Unity Catalog
• use WH tags to facilitate observability
• use API/Terraform to automate WH provisioning
• use API (i.e. for query, alerts) to monitor query executions and for QA purposes.
• use serverless warehouses for lower admin overhead and faster scaling of interactive/bi workloads

[Anonymous · Accepted solution]

Hi @Phani1 

We haven't heard from you since the last response from @Vinay_M_R , and I was checking back to see if her suggestions helped you.

Or else, If you have any solution, please share it with the community, as it can be helpful to others. 

Also, Please don't forget to click on the "Select As Best" button whenever the information provided helps resolve your question.