hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sergey Shelukhin (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-12354) MapJoin with double keys is slow on MR
Date Thu, 05 Nov 2015 23:02:27 GMT

    [ https://issues.apache.org/jira/browse/HIVE-12354?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14992666#comment-14992666
] 

Sergey Shelukhin commented on HIVE-12354:
-----------------------------------------

Taken from Arrays hashCode. Repeated overflow may produce more variability in the low bits,
I dunno.
Tested it locally btw, the code does execute on double join at least :)

> MapJoin with double keys is slow on MR
> --------------------------------------
>
>                 Key: HIVE-12354
>                 URL: https://issues.apache.org/jira/browse/HIVE-12354
>             Project: Hive
>          Issue Type: Bug
>            Reporter: Sergey Shelukhin
>            Assignee: Sergey Shelukhin
>         Attachments: HIVE-12354.patch
>
>
> Double keys are also a common type when comparing numbers with strings, and such, so
it happens more often than one would expect. This is due to HADOOP-12217



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message