Repository navigation
Add queue benchmarks and record results - #235
ianmacartney wants to merge 9 commits into
Conversation
|
Warning This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
This stack of pull requests is managed by Graphite. Learn more about stacking. |
commit: |
0fa5206 to
420c1cf
Compare
There was a problem hiding this comment.
All reported issues were addressed
Not reviewed (too large): example/benchmarks/2026-09-08.json (~5,865 lines) - if these are generated or fixture files, add them to ignored paths to exclude them from future reviews.
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
There was a problem hiding this comment.
All reported issues were addressed across 16 files (changes from recent commits).
Not reviewed (too large): example/benchmarks/2026-09-09/results.json (~7,332 lines) - if these are generated or fixture files, add them to ignored paths to exclude them from future reviews.
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
There was a problem hiding this comment.
All reported issues were addressed across 7 files (changes from recent commits).
Tip: Review your code locally with the cubic CLI to iterate faster.
Re-trigger cubic
387a3ae to
a9d1527
Compare
25b49a4 to
5ca4e83
Compare
a9d1527 to
24b91d6
Compare
5ca4e83 to
49c4e3a
Compare
24b91d6 to
95b2ccd
Compare
f85cd47 to
e8216bb
Compare
95b2ccd to
5b038c1
Compare
e8216bb to
39f7ce3
Compare
5b038c1 to
a506b21
Compare
39f7ce3 to
e453a2f
Compare
a506b21 to
215149a
Compare
e453a2f to
5352d16
Compare
215149a to
ee5640c
Compare

Add queue latency/scheduling probes and reproducible benchmarks for baseline, timestamp ordering, and packed pending starts. Record task outcomes and package versions, remove unused probes, and fix cleanup skipping tied creation timestamps. Validate run settings, logging, and enqueue coverage before accepting results.
September 9 median paired throughput changes:
Packing reduced enqueue writes by 49.5%. Its mutation throughput gain was variable (-4.9% to +10.3%). Full results and methodology include the cleanup interruption and all measured revisions.
Validation: 183,000 successful benchmark tasks; 175 workpool tests and 15 benchmark-tooling tests pass; both run in CI. Build, typecheck, lint, dashboard build, formatting, and Python checks pass.