NikhilTarte-Arm opened a new pull request, #13649:
URL: https://github.com/apache/gluten/pull/13649

   ## What changes are proposed in this pull request?
   
   After #11768, gluten-it reports only query time because the previous 
planning-time measurement was inaccurate. It summed `QueryPlanningTracker` rule 
durations, which do not represent elapsed planning time: nested rules can 
overlap, while planning work outside tracked rules can be omitted.
   
   This pull request restores separate planning and query timing using a single 
monotonic clock with explicit, non-overlapping boundaries. Planning Time covers 
`spark.sql` through initial physical planning and SQL execution setup. Query 
Time covers result collection, including runtime preparation, AQE replanning 
and subquery execution.
   
   Planning and collection remain within one SQL execution to preserve the 
execution ID, listener notifications, error handling and row conversion. Both 
timings are reported and aggregated in query, comparison and parameterized 
results, and the timing boundaries are documented.
   
   ## How was this patch tested?
   - Manually verified the planning and collection timing split with AQE 
enabled and disabled, including result aggregation and listener cleanup.
   - Verified one SQL execution ID, Gluten UI reporting, row conversion and 
query-failure propagation with Spark 3.4.4, 3.5.5, 4.0.2 and 4.1.1 on x86_64 
with JDK 17 and 21.
   
   ## Was this patch authored or co-authored using generative AI tooling?
   
   Generated-by: OpenAI Codex gpt-6-astra


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to