Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

Thanks Sam for your investigation. I revisited the logs and confirmed that the JDK has never changed. 'java -version' get: > openjdk version "11.0.19" 2023-04-18 LTS > OpenJDK Runtime Environment (Red_Hat-11.0.19.0.7-2) (build 11.0.19+7-LTS) > OpenJDK 64-Bit Server VM (Red_Hat-11.0.19.0.7-2)

Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

After completing the side quest [1] of enabling async profiler when running the JMH benchmarks I've been unable to reproduce the performance change between the last known good run and the first run highlighted as a regression. Results from my

Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

> I guess that improvement is a fluctuation. You can double check the performance results[1] of the last few days. The performance isn't recovered. Hmm yeah the improvement was a fluctuation and smaller than I remembered seeing (maybe I had zoomed into the timeline too much). > I fixed an issue

Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

Thanks Sam for the comment! > It looks like the most recent run of JDK 11 saw a big improvement of the > performance of the test. I guess that improvement is a fluctuation. You can double check the performance results[1] of the last few days. The performance isn't recovered. > That improvement

Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

It looks like the most recent run of JDK 11 saw a big improvement[1] of the performance of the test. That improvement seems related to [2] which is a fix for FLINK-35215 [3]. That suggests to me that the test isn't as isolated to the performance of the code its trying to test as would be ideal.

Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

Thanks Piotr and Zakelly for your advice. I have also tested this in my local machine, and I can not reproduce it even if switch to JDK11. This must have something to do with the benchmark's testing environment. > I will update FLINK-35040[1] to "Won't fix" and update the Priority from Blocker

Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

Hi Piotr and Zakelly, Thanks a lot for providing these historical JIRAs related to serializerHeavyString performance. It seems this similar issue has happened multiple times. And thanks for clarifying the principle of benchmark. I have encountered the semi-stable state for other benchmarks as

Re: [DISCUSS] The performance of serializerHeavyString starts regress since April 3

Hi Rui and RMs of Flink 1.20, Thanks for driving this! Available information indicates this issue is environment- and JDK-specific, and I also failed to reproduce it in my Mac. Thus I guess it is caused by JIT behavior, which is unpredictable and vulnerable to disturbance of the codebase.

[DISCUSS] The performance of serializerHeavyString starts regress since April 3

Hi devs: We(release managers of flink 1.20) wanna update one performance regresses to the flink dev mail list. # Background: The performance of serializerHeavyString starts regress since April 3, and we created FLINK-35040[1] to follow it. In brief: - The performance only regresses for jdk 11,