hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nigel Daley (JIRA)" <j...@apache.org>
Subject [jira] Updated: (MAPREDUCE-2079) PipesMapRunner does not use map output classes
Date Mon, 10 Jan 2011 20:55:47 GMT

     [ https://issues.apache.org/jira/browse/MAPREDUCE-2079?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Nigel Daley updated MAPREDUCE-2079:
-----------------------------------

      Priority: Major  (was: Blocker)
    Issue Type: Improvement  (was: Bug)

Not clear why this is a blocker.  Also, seems like an improvement, not a bug.  Marking it
as such.

> PipesMapRunner does not use map output classes
> ----------------------------------------------
>
>                 Key: MAPREDUCE-2079
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2079
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: pipes
>    Affects Versions: 0.20.1, 0.20.2, 0.20.3, 0.21.0, 0.21.1, 0.22.0
>         Environment: Ubuntu 10.04 Lucid 64bit, Java(TM) SE Runtime Environment (build
1.6.0_20-b02), Hadoop 20.2 (Cloudera distribution)
>            Reporter: Cosmin Catanoaie
>   Original Estimate: 0.17h
>  Remaining Estimate: 0.17h
>
> In org.apache.hadoop.mapred.pipes.PipesMapRunner.java at lines 71-72 instead of:
>  - job.getOutputKeyClass()    and
>  - job.getOutputValueClass()
> there should be called:
>  - job.getMapOutputKeyClass()    and
>  - job.getMapOutputValueClass()
> This modification will allow pipe jobs to use intermediate map output classes different
from the job final output classes.

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


Mime
View raw message