I'm working on Databricks Business Dashboards and trying to calculate interactive cluster compute time and total dollar spend per workspace.

As per standard understanding, the total dollar spent = Interactive Clusters + Job Clusters + SQL Warehouses.

I'm able to fetch total spend using system, billing, usage and price references from system ,billing ,list prices. However, to isolate Interactive Cluster usage only, is there any Databricks-recommended filter (e.g., cluster type, tags, or system fields) that cleanly segregates Interactive Clusters from Job or SQL clusters?

Appreciate any documented best practices or SQL-level filtering logic you've used successfully.