Rangsh commented on issue #12063: URL: https://github.com/apache/seatunnel/issues/12063#issuecomment-5585499076
Thanks @nzw921rx — understood. I will follow the same evidence pattern as #12176: 1. Use **Benchmarks Diagnostics** (on the fork, since I cannot dispatch it on `apache/seatunnel` via API) for the growth methods. 2. Download the workflow artifacts for local analysis (GC / JFR / profiles). 3. Present a **complete causal chain**: before profiles showing the suspected fixture cost (`FileMapStore.loadAll` / full WAL replay on iteration tear-down), then after profiles / scores showing that cost is removed, with Score/Error/CV on the same settings. I am starting with: - before: `dev` - after: PR #12173 (`improve/zeta-job-lifecycle-growth-stability-12063`) - method: `IMapJobStorageBenchmark.runningJobGrowth$` (`initialStoredJobCount=0`), GC + JFR first - then expand to `completedJobHistoryGrowth` and `initialStoredJobCount=1000` once the first artifacts are in I will post screenshots / artifact links and the causal write-up here as soon as the first diagnostics products are available. If anything about the workflow form is unclear, I will screenshot and ask immediately as you suggested. -- 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]
