[jira] [Updated] (PIG-3365) Run as uber job if there is only one input split

["Lorand Bendig (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/PIG-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Lorand Bendig updated PIG-3365:
-------------------------------

    Attachment: PIG-3365.patch

Patch based on the comments above. When {{mapreduce.job.ubertask.maxbytes}} is not set, default block size is the block size of the job's staging dir file system. 
(similarly what Hadoop 2.4.0 does in {{JobImpl#makeUberDecision}}). I renamed the property to pig.opt.ubertask.hint and set it to true by default.

I can change {{opt.fetch}} to {{pig.opt.fetch}} but changing the others would likely break backward compatibility.


> Run as uber job if there is only one input split
> ------------------------------------------------
>
>                 Key: PIG-3365
>                 URL: https://issues.apache.org/jira/browse/PIG-3365
>             Project: Pig
>          Issue Type: Improvement
>            Reporter: Rohini Palaniswamy
>            Assignee: Lorand Bendig
>              Labels: Performance
>         Attachments: PIG-3365.patch
>
>
> Hadoop 2 has support for uber mode (mapreduce.job.ubertask.enable=true) which runs the map and reduce on Application Master itself and reduces the overhead of launching a separate map/reduce task. 



--
This message was sent by Atlassian JIRA
(v6.2#6252)