Databricks Query Studio

Query Studio gives you visibility into the queries running against your Databricks clusters — historical and real-time — so you can find expensive or long-running queries and act on them, including aborting a query directly from ADOC.

Similar queries are grouped using fingerprinting: queries with the same structure share a fingerprint ID, giving you an aggregate view of how often a query pattern runs and what it costs, rather than reviewing every execution individually.

Filters

The Data Source Filter switches between connected Databricks accounts or projects.

Additional filters, available in the left pane, are contextual — for example, selecting a cluster narrows the database filter to that cluster's databases. A global calendar filter controls the time range reflected in the heatmap. Select refresh to fetch the latest data from Databricks.

Filter	Description
Execution Status	Completed, failed, or in progress.
Query Type	SQL, Python, or another execution type.
User Name	The user who initiated the query.
Cluster Name	The cluster the query ran on.
Notebook Path	The file path of the notebook the query ran from.
Cluster Type	All-purpose, job-specific, or another cluster category.

Viewing query data

Query Studio offers three ways to view query data, plus a per-user grouping.

Query Aggregate table

Groups similar queries into a query fingerprint. Applying a left-pane filter refreshes this table accordingly.

Column	Description
Sample Query	The representative query text. Select the copy icon to copy it, or select the query to open the Query Details page.
Distinct Warehouses	The number of unique warehouses the fingerprint's queries ran on.
Query Count	The number of queries grouped under this fingerprint.
Cloud Services Credits	Total credits consumed by the fingerprint.
Query Cost	Estimated resource cost of the fingerprint.
Avg / Max / Median Execution Time	Execution time statistics across the fingerprint's queries.
Median Queued Provisioning Time	The median time queries in this fingerprint spent queued for provisioning.
Avg Compilation Time	Average compilation time across the fingerprint's queries.
Avg Total Elapsed Time	Average total time elapsed, across the fingerprint's queries.

You can filter to the top 50 queries in categories like Top 50 Expensive Queries or Top 50 Long Running Queries. Selecting a category updates the left-pane filters to match, and results are limited to that category's top 50.

No Grouping table

Shows every individual query with no fingerprinting applied.

Column	Description
Query Text	The full query. Select the copy icon to copy it.
Query Id	The query's unique identifier.
Execution Status	The query's execution status.
User Name	Who executed the query.
Query Cost	Estimated execution cost.
Credits Used Cloud Services	Cloud service credits consumed.
Warehouse Name	The warehouse the query ran on. Select it to open the Query Details page, including actual cost incurred.
Total Elapsed Time	Total time from queue to completion.
Start Time / End Time	When the query started and ended.
Bytes Scanned	Total data scanned by the query.
Rows Produced	Rows created by the query.
Warehouse Size	The size of the warehouse the query ran on.
Query Type	The operation performed — for example, Select or Update.
Database Name / Schema Name	The database and schema the query ran against.
Execution Time / Compilation Time	Time spent executing and compiling the query.
Partitions Scanned / Partitions Total / % Partitions Scanned	Partition scan coverage for the query.
Error Code / Error Message	Details on any error encountered.
Queued Provisioning Time / Queued Repair Time / Queued Overload Time	Time spent queued, by cause.
Transaction Blocked Time	Time the query spent blocked by a transaction.
Bytes Spilled to Remote Storage / Local Storage	Data spilled to disk during execution.

Group by users

Groups results by user, showing Query Count, Avg / Max / Median Execution Time, Total Credits, and Query Cost per user — useful for identifying which users are driving the heaviest query load.

Exporting data

Select Export Raw Data to download the current view as a CSV, reflecting whatever filters are currently applied.

Query Detail view

Select a query in the No Grouping table to open its Query Detail view.

Metrics

Field	Description
Status	The query's execution result.
# Similar Queries	The number of queries similar to this one.
# Avg Schedule Frequency	[Placeholder — definition unconfirmed in source material. Confirm before publishing.]
Avg. Time	The query's average execution time.
Current Query Cluster	The cluster the query is currently associated with.
Cluster Count	The number of clusters used by the query.
# of Queries Notebook	Total queries in the same notebook.

Summary

Field	Description
Query ID	The query's unique identifier.
Cluster ID	The cluster used.
User Name	Who executed the query.
Query Fingerprint	The unique hash representing the query's structure.
Query Type	The SQL operation type — for example, SELECT.
Notebook Path	The notebook's file path.

Query shows the full query text, its execution status, and the query ID.

Compare Query shows how the current query compares to the best- and worst-performing queries on a chosen parameter, such as Execution Time — including the percentile by which the best query outperforms the current one, and by which the current query outperforms the worst. Use Add Query to compare against a specific query instead.

What's next

Databricks Compute – See how query activity relates to overall cluster and job performance.
Databricks Cost – Track the cost impact of the queries you identify here.

Last updated on

Was this page helpful?