AnalyticDB:Use query monitoring charts and SQL queries

Query monitoring charts

The Query Monitoring Information section visualizes query execution times across a selected time range. Each colored bar represents a single query — the longer the bar, the longer the query ran. Colors are assigned arbitrarily and carry no additional meaning.

Hover over a bar to see the query's start time, end time, and amount of scanned data. Click Details to open the query details page, where you can inspect the Query Properties, SQL Statement, and Execution Plan. For more information, see Use execution plans to analyze queries.

The chart displays up to 10,000 queries. It reflects only general search results — for example, queries matching Last 5 Minutes and Top 100 Most Time-consuming Queries. Advanced search filters applied in the SQL Queries tab do not affect the chart.

For an explanation of general and advanced search options, see the Search methods section in the SQL diagnostics overview.

SQL queries

The SQL Queries table shows more detail than the chart — including the full SQL statement, source address, and a breakdown of where time was spent. Use Show Advanced Search to narrow results by resource group, source address, database, or other criteria. Advanced search filters apply only to the table; they do not affect the monitoring chart.

Parameters

Parameter	Description
Actions	Click Diagnose to open the query details page. The page shows the Query Properties, SQL Statement, and Execution Plan tabs. For more information, see Use execution plans to analyze queries.
SQL	The executed SQL statement. Click the icon to copy it. Statements longer than 5,120 characters are truncated in this column — view the full statement in the downloaded Excel file or on the SQL Statement tab of the query details page.
Started At	The time when the query was submitted.
Query ID	The unique identifier of the query.
Source Address	The IP address of the client that submitted the query. To view the source address, go to Diagnostics and Optimization > Connection Information.
Database	The name of the database the query ran against. Use advanced search to filter by database.
Status	The execution status of the SQL statement. Valid values: Completed, Running, or Failed.
Total Duration	The total time from SQL submission to execution completion. Total Duration = Queuing Time + Execution Plan Duration + Execution Duration.
Queuing Time	The time from submission to an access node until execution starts. AnalyticDB for MySQL applies throttling on access nodes when concurrent query volume is high. If submitted queries reach the concurrency limit, new queries wait in queue. To avoid long queuing times: if resource consumption is low, increase the concurrency limit — see Priority queues and concurrency control of interactive resource groups. If resource consumption is already high, identify and address resource-heavy queries first. For more information about access nodes, see Overall architecture.
Execution Plan Duration	The time to generate an execution plan after the SQL statement reaches an access node. Long plan generation is usually caused by a complex SQL statement (such as excessive joins) or high resource usage on the access node. To resolve this, simplify the SQL statement or check for large result sets being returned to the client. For more information, see Query-level diagnosis results.
Execution Duration	The time to execute the SQL statement. Sort by this column to identify the most resource-intensive queries.
Returned Data	The amount of data returned to the client by the SELECT statement. Queries that return large amounts of data consume significant resources and can slow down other concurrent queries. Sort by this column to find queries returning excessive data.
Username	The name of the user that submitted the query. To view the username and the number of user connections, go to Diagnostics and Optimization > Connection Information.
Resource Group	The resource group that executed the SQL statement.
Peak Memory	The peak memory used by the query. AnalyticDB for MySQL executes queries in stages — dependent stages run serially, independent stages run in parallel — so memory usage peaks at the most resource-intensive point during execution.
Scanned Data	The amount of data read from the storage layer to the computing layer. More scanned data means higher disk I/O and greater demand on the computing layer, both of which slow queries down.
Total Stages	The total number of stages generated for the query. A large stage count indicates a complex SQL statement that requires extensive network coordination. If a query generates many stages, simplify the SQL statement. For background on stages, see Query process and execution plan.
ETL Written Rows	The number of rows written by an extract-transform-load (ETL) query.

What's next

• Use execution plans to analyze queries

• Query-level diagnosis results

• Priority queues and concurrency control of interactive resource groups