rkhachatryan commented on PR #19448: URL: https://github.com/apache/flink/pull/19448#issuecomment-1163264334
I've compared performance results. The top regressions are: ``` org.apache.flink.benchmark.MultipleInputBenchmark.multiInputOneIdleMapSink 15.28% org.apache.flink.benchmark.InputBenchmark.mapRebalanceMapSink 11.73% org.apache.flink.benchmark.MultiInputCheckpointingTimeBenchmark.checkpointMultiInput 7.01% org.apache.flink.benchmark.WindowBenchmarks.globalWindow 5.72% org.apache.flink.benchmark.MemoryStateBackendBenchmark.stateBackends 3.84% org.apache.flink.benchmark.SortingBoundedInputBenchmarks.sortedMultiInput 3.59% org.apache.flink.benchmark.MemoryStateBackendBenchmark.stateBackends 3.58% org.apache.flink.benchmark.TwoInputBenchmark.twoInputMapSink 3.42% org.apache.flink.benchmark.WindowBenchmarks.tumblingWindow 3.29% org.apache.flink.benchmark.AsyncWaitOperatorBenchmark.asyncWait 2.42% ``` Those are either caused by noisy benchmarks (e.g. `multiInputOneIdleMapSink`) and lie inside the usual bounds; or - probably - by a recent regression. So I'd suggest to wait for the results of the investigation of the latter, and then re-do the benchmark. -- 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]
