Rangsh commented on issue #12063: URL: https://github.com/apache/seatunnel/issues/12063#issuecomment-5653325397
@SEZ9 thanks — answering your three follow-ups. Also noting on #12173 that the spotless wrap is fixed on head `7948e2640` (`Code style` + `benchmark-test` JDK 8/11 green). **1. Same-worker run** I still cannot dispatch Benchmarks on `apache/seatunnel` (403). @nzw921rx (or another committer), could you please run **Benchmarks** there with `pr_number=12173` (`baseline → PR → PR → baseline`)? Once that finishes I will post the Score / Error / CV table for `IMapJobStorageBenchmark.(runningJobGrowth|completedJobHistoryGrowth)$` at `initialStoredJobCount=0` and `1000` from that single-host run. **2. Commit clarification** You were right to flag the mismatch. Correction: - My earlier note saying after run https://github.com/Rangsh/seatunnel/actions/runs/34354315647 was PR head `909f878` / table `c0d59da` was wrong. - Re-checked both fork dispatches: https://github.com/Rangsh/seatunnel/actions/runs/34354309911 and https://github.com/Rangsh/seatunnel/actions/runs/34354315647 both ran with **Checkout PR / Build PR skipped** (baseline-only on `dev` @ `75fd4ed`). So that Java 11 table is **not** a PR-head before/after — only cross-host observational noise (as already caveated with EPYC 9V74 vs 7763). - Current #12173 head for any same-worker run should be **`7948e2640`** (spotless fix on top of `c553a8c04`). **3. `runningJobGrowth` at `initialStoredJobCount=1000`** Agree 19.13% CV on that observational after host is still high. After the same-worker apache run: if CV stays in that band I will state explicitly whether I plan more forks/iterations or treat that fixture/count as inherently noisy, so the remaining variance is not over-read. Prefer not to guess until we have the single-host numbers. Diagnostics remain causal-only; Score/Error/CV waits on the same-worker Benchmarks run above. -- 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]
