hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Remus Rusanu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-5653) Vectorized Shuffle Join produces incorrect results
Date Sun, 27 Oct 2013 13:45:30 GMT

    [ https://issues.apache.org/jira/browse/HIVE-5653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13806333#comment-13806333
] 

Remus Rusanu commented on HIVE-5653:
------------------------------------

Cause is the VectorReduceSinkOperator, which does not initialize the tagBytes from the conf
during initializeOp. This results in reduce side ExecReducer.reduce reading all keys as tag
(==alias) 0. The reduce join operator sees all data in one side, nothing on the other, no
rows match.

> Vectorized Shuffle Join produces incorrect results
> --------------------------------------------------
>
>                 Key: HIVE-5653
>                 URL: https://issues.apache.org/jira/browse/HIVE-5653
>             Project: Hive
>          Issue Type: Sub-task
>            Reporter: Remus Rusanu
>            Assignee: Remus Rusanu
>
> Vectorized shuffle join should work out-of-the-box, but it produces empty result set.
Investigating.



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message