Skip to main content
Version: 6.2

Database Observability: Queries

The Database Observability: Queries dashboard identifies queries with high cumulative cost, regressions in execution time, errors, and high WAL volume.

Who it is for: database administrators, application developers, and performance specialists.

Data sources: db_statements*, db_activity*, db_logs*.

Main Sections​

1. Load Summary​

Shows the overall query profile:

  • execution count, unique query count, and QPS
  • average execution time and error count
  • distribution of queries by execution count and WAL volume

Query load summary

2. Queries by Wait Time​

Shows queries with the highest cumulative wait time and the distribution of time by wait type.

Queries by wait time

3. Query Analysis​

Shows normalized queries, their cumulative cost, execution count, average time, and signs of regression. Selecting a row in this section opens hidden panels with query execution records from the log and a card for the selected query. If an auto_explain plan is available, you can use it for further diagnosis.

Query analysis and execution records in the log

Selected query details

4. Database Errors​

Shows the distribution of errors by code and message, along with the original log events for the selected period.

Database errors