[
https://issues.apache.org/jira/browse/HIVE-16328?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15948689#comment-15948689
]
Rui Li commented on HIVE-16328:
-------------------------------
Hi [~csun], this is off the topic here, but IIRC, HoS uses raw data size (if
any) for map join, and HoMR uses file size for map join. Now the property
{{hive.spark.use.file.size.for.mapjoin}} seems to suggest it can make HoS also
use file size for map join. But in fact that's not the case, right?
> HoS: more aggressive mapjoin optimization when
> hive.spark.use.file.size.for.mapjoin is true
> -------------------------------------------------------------------------------------------
>
> Key: HIVE-16328
> URL: https://issues.apache.org/jira/browse/HIVE-16328
> Project: Hive
> Issue Type: Bug
> Components: Logical Optimizer
> Reporter: Chao Sun
> Assignee: Chao Sun
> Attachments: HIVE-16328.1.patch
>
>
> In HIVE-15489, when {{hive.spark.use.file.size.for.mapjoin}} is set to true,
> and if the JOIN op has any upstream RS operator, then we will stop converting
> the JOIN op to MAPJOIN op.
> However, this is overly conservative. A better solution is to treat the
> branch that has upstream RS as the big table and check if all other branches
> are map-only AND can fit in hash table size.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
