[
https://issues.apache.org/jira/browse/SPARK-32399?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Cheng Su updated SPARK-32399:
-----------------------------
Description: Currently for SQL full outer join, spark always does a sort
merge join no matter of how large the join children size are. Inspired by
recent discussion in
[https://github.com/apache/spark/pull/29130#discussion_r456502678] and
[https://github.com/apache/spark/pull/29181], I think we can support full outer
join in shuffled hash join in a way that - when looking up stream side keys
from build side {{HashedRelation}}. Mark this info inside build side
{{HashedRelation}}, and after reading all rows from stream side, output all
non-matching rows from build side based on modified {{HashedRelation}}. (was:
Currently for SQL full outer join, spark always does a sort merge join no
matter of how large the join children size are. Inspired by recent discussion
in [https://github.com/apache/spark/pull/29130#discussion_r456502678] and
[https://github.com/apache/spark/pull/29181], I think we can support full outer
join in shuffled hash join and broadcast hash join in a way that - when looking
up stream side keys from build side {{HashedRelation}}. Mark this info inside
build side {{HashedRelation}}, and after reading all rows from stream side,
output all non-matching rows from build side based on modified
{{HashedRelation}}. But more design details need to be figured out for this
JIRA.)
> Support full outer join in shuffled hash join
> ---------------------------------------------
>
> Key: SPARK-32399
> URL: https://issues.apache.org/jira/browse/SPARK-32399
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 3.1.0
> Reporter: Cheng Su
> Priority: Minor
>
> Currently for SQL full outer join, spark always does a sort merge join no
> matter of how large the join children size are. Inspired by recent discussion
> in [https://github.com/apache/spark/pull/29130#discussion_r456502678] and
> [https://github.com/apache/spark/pull/29181], I think we can support full
> outer join in shuffled hash join in a way that - when looking up stream side
> keys from build side {{HashedRelation}}. Mark this info inside build side
> {{HashedRelation}}, and after reading all rows from stream side, output all
> non-matching rows from build side based on modified {{HashedRelation}}.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
