DanielLeens commented on issue #12058: URL: https://github.com/apache/seatunnel/issues/12058#issuecomment-5584886339
Thanks for running the requested baseline reproduction against `origin/dev` rather than attributing the result to #12081. That makes the observation useful: the original `checkpointOverviewIncrementalUpdate` variance is present on the stated M1/JDK 11 environment, but one host/run cannot yet isolate JIT, GC, storage, or scheduling as the cause and it does not validate a production optimization. Please attach the raw JMH JSON and the complete per-fork iteration/GC evidence for this baseline, then add an A/A comparison on the same runner before testing any candidate change. Keep #12081 separate: its in-progress Build and its RequestFuture/WAL/HdfsWriter correctness scope do not establish that the checkpoint benchmark variance is resolved. The next conclusion should identify a measurable mechanism and a narrowly scoped experiment, not a throughput claim. -- 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]
