A visual account of how a query ran, covering operators, time taken, rows and peak memory, used to track down bottlenecks. Viewing it requires owning the query or CAN MONITOR on the warehouse, and on serverless it stands in for the Spark UI.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Query profile in context, with comparison tables and the common traps.
Terms in this definition
- CAN MONITOR
Meant for power users, this permission on a SQL warehouse is stronger than CAN USE but weaker than CAN MANAGE. It allows running queries and watching the warehouse, for example through its history of queries and their profiles.
- Warehouse
Fabric item offering complete T-SQL support (DML, DDL, multi-table transactions) over Delta tables held in OneLake; lakehouse SQL analytics endpoints, by contrast, are read-only.
- Serverless
Compute tier for single Azure SQL databases that scales automatically, pauses when idle and charges by the second. It is offered in General Purpose and Hyperscale, not Business Critical, and reserved capacity does not apply.
- Spark UI
Classic compute's Apache Spark web console. Databricks advises checking the jobs timeline first, then the slowest stage, skew or spill, I/O and lastly the SQL DAG. Serverless and SQL warehouses offer a query profile in its place.
Related terms
- Query cache
Repeated queries in Databricks SQL may be served from a local or remote result cache. Entries expire after 24 hours or as soon as a source table changes, and such cached answers come without a query profile.