databricks-metric-views

Standardize business KPIs in Databricks by defining governed metric views in YAML. This separates measure expressions from dimensions, ensuring consistent aggregations and preventing duplicated SQL logic across BI tools.

Create governed metric views in Databricks Unity Catalog by defining aggregated measures and display dimensions in YAML. Query them using MEASURE() semantics in your SQL warehouse to retrieve standardized data.

Yes, YAML supports defining complex aggregations for Databricks SQL, including ratios, filtered measures, distinct counts, and window measures like running totals, trailing windows, and YTD calculations for robust dimensional modeling.

Yes, managing metric views with YAML requires Databricks Runtime 17.2 or later for YAML v1.1 support. You also need appropriate SQL warehouse permissions, including SELECT, CREATE TABLE, and USE SCHEMA.

The best way to pre-compute aggregations for Databricks metric views is to enable optional materialization. This stores pre-computed results in Unity Catalog, optimizing query performance for heavy analytical workloads.

Grant access to Databricks metric views defined in YAML using MCP tooling. This allows you to manage permissions alongside creating, altering, describing, and dropping metric views within Unity Catalog.