dongjoon-hyun commented on a change in pull request #27078: [SPARK-30409][SPARK-29173][SQL][TESTS] Use `NoOp` datasource in SQL benchmarks URL: https://github.com/apache/spark/pull/27078#discussion_r365611089
########## File path: sql/core/benchmarks/JoinBenchmark-results.txt ########## @@ -2,74 +2,74 @@ Join Benchmark ================================================================================================ -OpenJDK 64-Bit Server VM 1.8.0_222-b10 on Linux 3.10.0-862.3.2.el7.x86_64 +OpenJDK 64-Bit Server VM 1.8.0_232-8u232-b09-0ubuntu1~18.04.1-b09 on Linux 4.15.0-1044-aws Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz Join w long: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative ------------------------------------------------------------------------------------------------------------------------ -Join w long wholestage off 4685 4814 182 4.5 223.4 1.0X -Join w long wholestage on 440 524 102 47.7 21.0 10.7X +Join w long wholestage off 4531 4557 37 4.6 216.1 1.0X +Join w long wholestage on 1214 1310 95 17.3 57.9 3.7X Review comment: @MaxGekk . For example, this case, this slowed down 3x times in both JDK8 and JDK11. Also, it's mostly consistent in the other cases in this suite. It's a huge regression which is orthogonal to this PR. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
