[
https://issues.apache.org/jira/browse/HIVE-18201?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16286566#comment-16286566
]
Gunther Hagleitner commented on HIVE-18201:
-------------------------------------------
[~ashutoshc]:
https://github.com/apache/hive/blob/22e7bffa1c1a9be1e8503ebe0f7adf39060d1979/ql/src/java/org/apache/hadoop/hive/ql/optimizer/ConvertJoinMapJoin.java#L108
We explicitly turn it off. The problem is that this decision needs to be cost
based (i.e. take a look at the total data transfer.). We ran into cases where
the xprod is much faster than map join...
> Disable XPROD_EDGE for sq_count_check() created for scalar subqueries
> ----------------------------------------------------------------------
>
> Key: HIVE-18201
> URL: https://issues.apache.org/jira/browse/HIVE-18201
> Project: Hive
> Issue Type: Bug
> Affects Versions: 3.0.0
> Reporter: Nita Dembla
> Attachments: query6.explain2.out
>
>
> sq_count_check() will either return an error at runtime or a single row. In
> case of query6, the subquery has avg() function that should return a single
> row. Attaching the explain.
> This does not need an x-prod, because it is not useful to shuffle the big
> table side for a cross-product against 1 row.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
