Rangsh commented on issue #12063: URL: https://github.com/apache/seatunnel/issues/12063#issuecomment-5541924536
Thanks @DanielLeens for confirming the claim and clarifying the investigation order. I will start from reproducible evidence first: run IMapJobStorageBenchmark.runningJobGrowth and IMapJobStorageBenchmark.completedJobHistoryGrowth with the same runner class, JDK, arguments, and storage configuration, then attach the JMH JSON plus CPU / wall / lock / GC or JFR profiles to identify whether the variance comes from the fixture, Hazelcast/MapStore path, serialization/allocation, or the execution environment. If the profiling points to a production bottleneck, I will keep the change narrowly scoped to that path, preserve running-job and completed-job state durability, add targeted behavior coverage, and include comparable before/after results. I will follow up with a focused PR and link it here once the analysis is ready. -- 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]
