[
https://issues.apache.org/jira/browse/ASTERIXDB-3331?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17796873#comment-17796873
]
ASF subversion and git services commented on ASTERIXDB-3331:
------------------------------------------------------------
Commit 6cba52b3d64696f4e60cb14617aded88baea430a in asterixdb's branch
refs/heads/master from Wail Alkowaileet
[ https://gitbox.apache.org/repos/asf?p=asterixdb.git;h=6cba52b3d6 ]
[ASTERIXDB-3331][COMP] Avoid pushing agg functions in columnar filter
- user model changes: no
- storage format changes: no
- interface changes: no
Details:
Aggregate functions should not be pushed down to data-scan
in columnar filter.
Change-Id: I6826279e24f92aad7ae7cd7de6b74a3811a57183
Reviewed-on: https://asterix-gerrit.ics.uci.edu/c/asterixdb/+/18011
Integration-Tests: Jenkins <[email protected]>
Tested-by: Jenkins <[email protected]>
Reviewed-by: Wail Alkowaileet <[email protected]>
Reviewed-by: Ali Alsuliman <[email protected]>
> Columnar filter should not pushdown aggregate functions
> -------------------------------------------------------
>
> Key: ASTERIXDB-3331
> URL: https://issues.apache.org/jira/browse/ASTERIXDB-3331
> Project: Apache AsterixDB
> Issue Type: Bug
> Components: COMP - Compiler
> Affects Versions: 0.9.9
> Reporter: Wail Y. Alkowaileet
> Assignee: Wail Y. Alkowaileet
> Priority: Major
> Fix For: 0.9.9
>
>
> Query:
> {noformat}SELECT VALUE d
> FROM ColumnDataset d
> WHERE EXISTS d.array{noformat}
> Plan:
> {noformat}distribute result [$$d] [cardinality: 0.0, op-cost: 0.0,
> total-cost: 0.0]
> -- DISTRIBUTE_RESULT |PARTITIONED|
> exchange [cardinality: 0.0, op-cost: 0.0, total-cost: 0.0]
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> select (neq(count($$d.getField(\"array\")), 0)) [cardinality: 0.0,
> op-cost: 0.0, total-cost: 0.0]
> -- STREAM_SELECT |PARTITIONED|
> project ([$$d]) [cardinality: 0.0, op-cost: 0.0, total-cost: 0.0]
> -- STREAM_PROJECT |PARTITIONED|
> exchange [cardinality: 0.0, op-cost: 0.0, total-cost: 0.0]
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> data-scan []<-[$$15, $$d] <- test.ColumnDataset filter on:
> neq(count($$d.getField(\"array\")), 0) [cardinality: 0.0, op-cost: 0.0,
> total-cost: 0.0]
> -- DATASOURCE_SCAN |PARTITIONED|
> exchange [cardinality: 0.0, op-cost: 0.0, total-cost: 0.0]
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> empty-tuple-source [cardinality: 0.0, op-cost: 0.0, total-cost:
> 0.0]
> -- EMPTY_TUPLE_SOURCE |PARTITIONED| {noformat}
>
> Filter *neq(count($$d.getField("array")), 0)* should not pushed to data-scan
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
