goutamadwant commented on issue #12060: URL: https://github.com/apache/seatunnel/issues/12060#issuecomment-5554857608
@DanielLeens @nzw921rx completed the benchmark-only changes and matched Linux A/A calibration - separated startup-inclusive timing from writes after coordinator activation and storage read-back - Java 8 and Java 11 each completed 96 fresh JVMs using identical artifacts for both comparison roles - verified raw results, log hashes and settings across 57,600 measured writes, with each JDK and timing contract analyzed separately - identical-code post-readiness ratios ranged from 0.951 to 1.216 on Java 8 and 0.921 to 1.046 on Java 11, so these results do not support a production speedup claim or a small regression threshold yet - production storage and cleanup are unchanged, and the full Build now passes The next step is to agree on the readiness boundary and confirmation budget before evaluating a production optimization. Coordinator activation and read-back do not mean all background startup or JIT activity has settled. [Java 8 calibration](https://github.com/goutamadwant/seatunnel/actions/runs/33983362872) [Java 11 calibration](https://github.com/goutamadwant/seatunnel/actions/runs/33983358973) [Passing Build](https://github.com/goutamadwant/seatunnel/actions/runs/33984408660) -- 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]
