Add per-slice and per-partition breakdowns to the query profiler - #22613
Add per-slice and per-partition breakdowns to the query profiler#22613prudhvigodithi wants to merge 4 commits into
Conversation
Signed-off-by: Prudhvi Godithi <pgodithi@amazon.com>
Signed-off-by: Prudhvi Godithi <pgodithi@amazon.com>
PR Reviewer Guide 🔍(Review updated until commit fb113e3)Here are some key observations to aid the review process:
|
PR Code Suggestions ✨Latest suggestions up to fb113e3 Explore these optional code suggestions:
Previous suggestionsSuggestions up to commit 5f31ef8
Suggestions up to commit d663f7e
|
Sample output |
|
❌ Gradle check result for d663f7e: FAILURE Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change? |
|
Very nice. Just to confirm, this will add I'm looking forward to see what visualization we can create out of this data. |
|
Persistent review updated to latest commit 5f31ef8 |
Its the slice's earliest operation started to when its latest one finished
Yes, this is true. The idea is get/debug the slowest slice which would further help with partition data with intra segment. |
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #22613 +/- ##
============================================
- Coverage 71.43% 71.43% -0.01%
- Complexity 76760 76823 +63
============================================
Files 6142 6142
Lines 357766 357880 +114
Branches 52148 52178 +30
============================================
+ Hits 255581 255657 +76
- Misses 81861 81890 +29
- Partials 20324 20333 +9 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
|
Thanks @asimmahmood1 I have updated to fix the tests will do some cleanup, adding @jainankitk @sohami |
Signed-off-by: Prudhvi Godithi <pgodithi@amazon.com>
|
Persistent review updated to latest commit fb113e3 |
Description
Adds a nested
slices[] → partitions[]breakdown to the query profiler, showing per-slice and (under intra-segment search) per-partition timings that were previously collapsed into only themax_/min_/avg_slice_*aggregates.This change adds per-slice/per-partition detail to the query section only (
profile.shards[].searches[].query[]). The same pattern can be extended, as follow-ups, to the collector and aggregation sections so the whole profile output shows slice/partition granularity consistently.Related Issues
Check List
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.