Bankim Bhavsar has posted comments on this change. ( http://gerrit.cloudera.org:8080/15683 )
Change subject: IMPALA-3741 [part 2]: Push runtime bloom filter to Kudu ...................................................................... Patch Set 3: (1 comment) http://gerrit.cloudera.org:8080/#/c/15683/3/common/thrift/PlanNodes.thrift File common/thrift/PlanNodes.thrift: http://gerrit.cloudera.org:8080/#/c/15683/3/common/thrift/PlanNodes.thrift@124 PS3, Line 124: BLOOM_MIN_MAX = 2 > Instead of 'are not sorted(partially sorted within each tablet)', I meant This is my understanding, I'll check further within Kudu team. When the predicate pushed to Kudu includes partition key columns then pruning using min-max filter will be done first eliminating bunch of files containing column values namely cfiles. When the predicate doesn't include partition key columns then all column values will need to be scanned and as per current implementation for every column value first Bloom filter will be checked before checking for range/min-max values[1]. So in case of non-partition key columns whether min-max filter will help depends upon whether min-max filter values are simply min and max of the values inserted in the Bloom filter or not derived from the values inserted in the Bloom filter. For the former case, I agree supplying min-max filter with Bloom filter won't help. [1] https://github.com/apache/kudu/blob/master/src/kudu/common/column_predicate.h#L332 -- To view, visit http://gerrit.cloudera.org:8080/15683 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I9100076f68ea299ddb6ec8bc027cac7a47f5d754 Gerrit-Change-Number: 15683 Gerrit-PatchSet: 3 Gerrit-Owner: Wenzhe Zhou <[email protected]> Gerrit-Reviewer: Aman Sinha <[email protected]> Gerrit-Reviewer: Bankim Bhavsar <[email protected]> Gerrit-Reviewer: Impala Public Jenkins <[email protected]> Gerrit-Reviewer: Thomas Tauber-Marshall <[email protected]> Gerrit-Reviewer: Wenzhe Zhou <[email protected]> Gerrit-Comment-Date: Fri, 10 Apr 2020 03:23:19 +0000 Gerrit-HasComments: Yes
