Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/11209#discussion_r72918850 --- Diff: sql/catalyst/src/test/scala/org/apache/spark/sql/HashBenchmark.scala --- @@ -119,11 +146,12 @@ object HashBenchmark { .add("map", mapOfInt) .add("mapOfMap", MapType(IntegerType, mapOfInt)) /* - Intel(R) Core(TM) i7-4960HQ CPU @ 2.60GHz + Intel(R) Core(TM) i7-4750HQ CPU @ 2.00GHz Hash For map: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative ------------------------------------------------------------------------------------------- - interpreted version 1820 / 1861 0.0 444347.2 1.0X - codegen version 205 / 223 0.0 49936.5 8.9X + interpreted version 1612 / 1618 0.0 393553.4 1.0X + codegen version 149 / 150 0.0 36381.2 10.8X + codegen version 64-bit 144 / 145 0.0 35122.1 11.2X */ --- End diff -- This result is because we have removed the `hashCode` implementation from `ArrayBasedMapData`, and the system default `hashCode` is really fast(almost no-op). We did this because they have same logic at the time, but it's not true now. @gatorsmile do you wanna send a PR to fix this(use interpreted hash expression instead of `BaseGenericInternalRow.hashCode`)? thanks!
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org