You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Namit Jain (JIRA)" <ji...@apache.org> on 2008/09/11 03:36:44 UTC

[jira] Created: (HADOOP-4158) [hive] redundant columns being passed in group by stage 1

[hive] redundant columns being passed in group by stage 1
---------------------------------------------------------

                 Key: HADOOP-4158
                 URL: https://issues.apache.org/jira/browse/HADOOP-4158
             Project: Hadoop Core
          Issue Type: Bug
          Components: contrib/hive
            Reporter: Namit Jain
            Assignee: Namit Jain


In the first map-reduce stage, all input columns are copied currently. Instead, we should walk over the filter and projections for that destination and only copy columns which occur there

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (HADOOP-4158) [hive] redundant columns being passed in group by stage 1

Posted by "Zheng Shao (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HADOOP-4158?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12630338#action_12630338 ] 

Zheng Shao commented on HADOOP-4158:
------------------------------------

Agree. This applies to the hash-based (multiple) group by I guess, because the sort-based (single) group by already pre-compute the target key and aggregation columns so only those that are necessary are passed.

> [hive] redundant columns being passed in group by stage 1
> ---------------------------------------------------------
>
>                 Key: HADOOP-4158
>                 URL: https://issues.apache.org/jira/browse/HADOOP-4158
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: contrib/hive
>            Reporter: Namit Jain
>            Assignee: Namit Jain
>
> In the first map-reduce stage, all input columns are copied currently. Instead, we should walk over the filter and projections for that destination and only copy columns which occur there

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.