You are viewing a plain text version of this content. The canonical link for it is here.

Posted to mapreduce-issues@hadoop.apache.org by "Jason Lowe (JIRA)" <ji...@apache.org> on 2015/11/24 23:05:11 UTC

[jira] [Updated] (MAPREDUCE-5883) "Total megabyte-seconds" in job counters is slightly misleading

     [ https://issues.apache.org/jira/browse/MAPREDUCE-5883?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jason Lowe updated MAPREDUCE-5883:
----------------------------------
       Resolution: Fixed
     Hadoop Flags: Reviewed
    Fix Version/s: 2.7.3
                   2.6.3
           Status: Resolved  (was: Patch Available)

Thanks, Nathan!  I committed this to trunk, branch-2, branch-2.7, and branch-2.6.

> "Total megabyte-seconds" in job counters is slightly misleading
> ---------------------------------------------------------------
>
>                 Key: MAPREDUCE-5883
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5883
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>    Affects Versions: 3.0.0, 2.4.0
>            Reporter: Nathan Roberts
>            Assignee: Nathan Roberts
>            Priority: Minor
>              Labels: BB2015-05-TBR
>             Fix For: 2.6.3, 2.7.3
>
>         Attachments: MAPREDUCE-5883.patch
>
>
> The following counters are in milliseconds so "megabyte-seconds" might be better stated as "megabyte-milliseconds"
> MB_MILLIS_MAPS.name=               Total megabyte-seconds taken by all map tasks
> MB_MILLIS_REDUCES.name=            Total megabyte-seconds taken by all reduce tasks
> VCORES_MILLIS_MAPS.name=           Total vcore-seconds taken by all map tasks
> VCORES_MILLIS_REDUCES.name=        Total vcore-seconds taken by all reduce tasks



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