Michael Smith has posted comments on this change. ( http://gerrit.cloudera.org:8080/21687 )
Change subject: IMPALA-5443: Apply codegen rows threshold per node ...................................................................... Patch Set 3: (1 comment) http://gerrit.cloudera.org:8080/#/c/21687/2/fe/src/main/java/org/apache/impala/util/PlanNodeCodegenVisitor.java File fe/src/main/java/org/apache/impala/util/PlanNodeCodegenVisitor.java: http://gerrit.cloudera.org:8080/#/c/21687/2/fe/src/main/java/org/apache/impala/util/PlanNodeCodegenVisitor.java@57 PS2, Line 57: long numRowsPerInstance = (long)Math.ceil(numRows / (double)numNodes); I think it would make sense for this visitor to take into account ProcessingCost or expression complexity. An simple expression that's trivial to codegen seems like it would be worth doing for fewer rows than a really complex expression that takes awhile to codegen. Although presumably the interpreted cost also goes up. Need to come up with a methodology for deciding the threshold and how query characteristics affect it. -- To view, visit http://gerrit.cloudera.org:8080/21687 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I1b34d4f2ef0d98fcd918a8a546acde70e09d18d7 Gerrit-Change-Number: 21687 Gerrit-PatchSet: 3 Gerrit-Owner: Michael Smith <[email protected]> Gerrit-Reviewer: Impala Public Jenkins <[email protected]> Gerrit-Reviewer: Michael Smith <[email protected]> Gerrit-Comment-Date: Wed, 21 Aug 2024 00:36:40 +0000 Gerrit-HasComments: Yes
