Monitor BI Engine
BigQuery BI Engine acceleratesBigQuery for BI scenarios using memory cache and faster execution.Acceleration details can be monitored usingINFORMATION_SCHEMA andCloud Monitoring metrics.
Cloud Monitoring
You can monitor and configure alerts for BigQuery BI Engine withCloud Monitoring. To learn how to create dashboard forBI Engine metrics, seeCreating charts.
The following metrics are provided for BigQuery BI Engine:
| Resource | Metric | Details |
|---|---|---|
| BigQuery Project | Reservation Total Bytes | Total capacity allocated to one Google Cloud project |
| BigQuery Project | Reservation Used Bytes | Total capacity used in one Google Cloud project |
| BigQuery Project | BI Engine Top Tables Cached Bytes | Cache usage per table. This metric displays the topN tables per region report usage. |
Query statistics for BI Engine
This section explains how to find query statistics to help monitor, diagnose, and troubleshoot BI Engine use.
BI Engine acceleration modes
With BI Engine acceleration enabled, your query can run in any oneof these four modes:
BI_ENGINE_DISABLED | BI Engine disabled the acceleration.biEngineReasons specifies a more detailed reason. The query was run using the BigQuery execution engine. |
PARTIAL_INPUT | Part of the query input was accelerated using BI Engine. Asdescribed inQuery optimization and acceleration, a query plan is generally broken down into multiple input stages. BI Engine supports the common typesof subquery patterns that are typically used in dashboarding. If the queryconsists of multiple input stages, only a few of which fall under thesupported use cases, then BI Engine runs the unsupported stagesusing the normal BigQuery engine without acceleration. In thissituation, BI Engine returns aPARTIALacceleration code, and usesbiEngineReasons to populate the reasonfor not accelerating other input stages. |
FULL_INPUT | All input stages of the query were accelerated usingBI Engine. Cached data is reused across queries, and the computationthat follows immediately after reading the data is accelerated. |
FULL_QUERY | The entire query was accelerated usingBI Engine. |
View BigQuery API job statistics
Detailed statistics on BI Engine are available through theBigQuery API.
To fetch the statistics associated with BI Engine acceleratedqueries, run the following bq command-line tool command:
bq show --format=prettyjson -j job_idIf the project is enabled for BI Engine acceleration, then theoutput produces a new field,biEngineStatistics. Here is a sample jobreport:
"statistics": { "creationTime": "1602175128902", "endTime": "1602175130700", "query": { "biEngineStatistics": { "biEngineMode": "DISABLED", "biEngineReasons": [ { "code": "UNSUPPORTED_SQL_TEXT", "message": "Detected unsupported join type" } ] },For more information about theBiEngineStatistics field, see theJob reference.
BigQuery information schema statistics
BI Engine acceleration statistics are included in theBigQueryINFORMATION_SCHEMAviews as part of theINFORMATION_SCHEMA.JOBS_BY_* views in thebi_engine_statistics column.For example, this query returns thebi_engine_statistics for all of thecurrent projects' jobs for the last 24 hours:
SELECT creation_time, job_id, bi_engine_statisticsFROM `region-us`.INFORMATION_SCHEMA.JOBS_BY_PROJECTWHERE creation_time > TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 1 DAY) AND job_type = "QUERY"Use the following format to specifyregionality for theproject-id,region, andviews in theINFORMATION_SCHEMA view:
`PROJECT_ID`.`region-REGION_NAME`.INFORMATION_SCHEMA.VIEW
View Looker Studio information schema details
You can track which Looker Studio reports and data sources areused by BigQuery by viewing theINFORMATION_SCHEMA.JOBSview. EveryLooker Studio query in BigQuery creates an entrywithreport_id anddatasource_id labels. Those IDs appear at the end ofLooker Studio URL when opening a report or data source page.For example, a report with URLhttps://lookerstudio.google.com/navigation/reporting/my-report-id-123 has areport ID of"my-report-id-123".
The following examples show how to view reports and data sources:
Find report and data source URL for each Looker Studio BigQuery job
-- Standard labels used by Looker Studio.DECLARErequestor_keySTRINGDEFAULT'requestor';DECLARErequestor_valueSTRINGDEFAULT'looker_studio';CREATETEMPFUNCTIONGetLabel(labelsANYTYPE,label_keySTRING)AS((SELECTl.valueFROMUNNEST(labels)lWHEREl.key=label_key));CREATETEMPFUNCTIONGetDatasourceUrl(labelsANYTYPE)AS(CONCAT("https://lookerstudio.google.com/datasources/",GetLabel(labels,'looker_studio_datasource_id')));CREATETEMPFUNCTIONGetReportUrl(labelsANYTYPE)AS(CONCAT("https://lookerstudio.google.com/reporting/",GetLabel(labels,'looker_studio_report_id')));SELECTjob_id,GetDatasourceUrl(labels)ASdatasource_url,GetReportUrl(labels)ASreport_url,FROM`region-us`.INFORMATION_SCHEMA.JOBSjobsWHEREcreation_time>TIMESTAMP_SUB(CURRENT_TIMESTAMP(),INTERVAL7DAY)ANDGetLabel(labels,requestor_key)=requestor_valueLIMIT100;
View jobs produced by using a report and data source
-- Specify report and data source id, which can be found at the end of Looker Studio URLs.DECLAREuser_report_idSTRINGDEFAULT'*report id here*';DECLAREuser_datasource_idSTRINGDEFAULT'*datasource id here*';-- Looker Studio labels for BigQuery.DECLARErequestor_keySTRINGDEFAULT'requestor';DECLARErequestor_valueSTRINGDEFAULT'looker_studio';DECLAREdatasource_keySTRINGDEFAULT'looker_studio_datasource_id';DECLAREreport_keySTRINGDEFAULT'looker_studio_report_id';CREATETEMPFUNCTIONGetLabel(labelsANYTYPE,label_keySTRING)AS((SELECTl.valueFROMUNNEST(labels)lWHEREl.key=label_key));SELECTcreation_time,job_id,FROM`region-us`.INFORMATION_SCHEMA.JOBSjobsWHEREcreation_time>TIMESTAMP_SUB(CURRENT_TIMESTAMP(),INTERVAL7DAY)ANDGetLabel(labels,requestor_key)=requestor_valueANDGetLabel(labels,datasource_key)=user_datasource_idANDGetLabel(labels,report_key)=user_report_idORDERBY1LIMIT100;
Cloud Logging
BI Engine acceleration is part of BigQuery jobprocessing. To inspect BigQuery jobs for a specific project,see theCloud Logging page with a payload ofprotoPayload.serviceName="bigquery.googleapis.com".
What's next
- Learn more aboutCloud Monitoring.
- Learn more about Monitoringcharts.
- Learn more about Monitoringalerts.
- Learn more aboutCloud Logging.
Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2026-02-18 UTC.