[
https://issues.apache.org/jira/browse/HIVE-26968?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17729297#comment-17729297
]
Seonggon Namgung commented on HIVE-26968:
-----------------------------------------
[~zabetak] If my memory serves me right, there is a slight difference in DPP
optimization between native and non-native (e.g. iceberg) tables. So I guess
that hive.optimize.shared.work.dppunion matters only when we reproduce this
issue on iceberg table.
Since the problem comes from SharedWorkOptimizer rather than Iceberg related
things, I think the new qfile is sufficient to test this issue.
> SharedWorkOptimizer merges TableScan operators that have different DPP parents
> ------------------------------------------------------------------------------
>
> Key: HIVE-26968
> URL: https://issues.apache.org/jira/browse/HIVE-26968
> Project: Hive
> Issue Type: Sub-task
> Affects Versions: 4.0.0-alpha-2
> Reporter: Seonggon Namgung
> Assignee: Seonggon Namgung
> Priority: Critical
> Labels: hive-4.0.0-must, pull-request-available
> Attachments: TPC-DS Query64 OperatorGraph.pdf
>
> Time Spent: 50m
> Remaining Estimate: 0h
>
> SharedWorkOptimizer merges TableScan operators that have different DPP
> parents, which leads to the creation of semantically wrong query plan.
> In our environment, running TPC-DS query64 on 1TB Iceberg format table
> returns no rows because of this problem. (The correct result has 7094 rows.)
> We use hive.optimize.shared.work=true,
> hive.optimize.shared.work.extended=true, and
> hive.optimize.shared.work.dppunion=false to reproduce the bug.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
