Event Based Query Analysis
The Event Based Query Analysis dashboard provides a high-level view of the impact of a selected event on query performance. It compares query execution metrics before and after the event date and highlights whether overall performance improved or degraded.
The dashboard includes the following filters:
Event Date – Select the event date to analyze its impact on query performance.
Days Before – Specify the number of days prior to the event date to include in the analysis.
Days After – Specify the number of days after the event date to include in the analysis.
The dashboard displays the following summary information:
Total Queries Executed Before Event Date – Total number of queries executed during the selected pre-event period.
Total Queries Executed After Event Date – Total number of queries executed during the selected post-event period.
Event Impact Summary – AI-generated summary describing the overall impact of the selected event on query performance.
Common Queries – Number of queries executed in both the pre-event and post-event periods.
Improved – Number of common queries whose performance improved after the event.
Degraded – Number of common queries whose performance degraded after the event.
Avg Query Performance Pre Event Date (Sec) – Average execution time of common queries before the event.
Avg Query Performance Post Event Date (Sec) – Average execution time of common queries after the event.
Difference (Sec) – Difference between the average query execution times before and after the event.
Detailed Query Analysis – Provides a detailed breakdown of query-level performance changes. Click More Info to view additional analysis.
Detailed Query Analysis
The Detailed Query Analysis section provides a query-level comparison of performance before and after the selected event. It helps identify which queries experienced performance improvements or degradations.
The following filters are available:
Semantic Model – Filters the analysis by selecting semantic models.
Query Source – Filters queries based on their execution source, such as cache, disk, query engine cache, or metadata.
The analysis table includes the following columns:
Query – The SQL query or query pattern analyzed.
Pre (Sec) – Average execution time before the event.
Post (Sec) – Average execution time after the event.
Diff (%) – Percentage change in execution time between the pre-event and post-event periods.
Status – Indicates whether the query performance Improved or Degraded after the event.
Use this section to identify specific queries affected by the event and evaluate the magnitude of the performance change.
Top Improved and Degraded Queries
The dashboard highlights the queries with the most significant performance changes after the selected event.
Top 5 Degraded Queries
Displays the five queries with the highest performance degradation. The table includes:
Query – Query text or query pattern.
Pre (Sec) – Average execution time before the event.
Post (Sec) – Average execution time after the event.
Diff (%) – Percentage increase in execution time after the event.
Top 5 Improved Queries
Displays the five queries with the greatest performance improvement. The table includes:
Query – Query text or query pattern.
Pre (Sec) – Average execution time before the event.
Post (Sec) – Average execution time after the event.
Diff (%) – Percentage reduction in execution time after the event.