hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Amar Kamat (JIRA)" <j...@apache.org>
Subject [jira] Commented: (MAPREDUCE-339) JobTracker should give preference to failed tasks over virgin tasks so as to terminate the job ASAP if it is eventually going to fail.
Date Fri, 03 Jul 2009 12:34:47 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-339?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12726943#action_12726943
] 

Amar Kamat commented on MAPREDUCE-339:
--------------------------------------

Also we might want to be greedy and schedule failed task over virgin tasks as soon as we find
out that the almost all the tasks for a job is failing. Something like 
{code}
if (num-attempts-scheduled > x% of total-tasks  && num-attempts-failed/num-attempts-scheduled
> y%) then
schedule failed tasks greedily
{code}

Thoughts?

> JobTracker should give preference to failed tasks over virgin tasks so as to terminate
the job ASAP if it is eventually going to fail. 
> ---------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-339
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-339
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>            Reporter: Gautam Kowshik
>            Assignee: Devaraj Das
>
> Case in point... I have 1585 maps and 160 slots (40 nodes). The job is such that all
maps fail within 2-3 minutes. The job takes forever to realise that the job is bad. It took
2526 failures for it to reach 4 failed attempts for a task. 
> As I understand, currently the JT prefers a failed task if and only if a task tracker
with a split replica for that map came asking for a task. In fact there may not be a single
TT at all in the mapred cluster which has a replica for the splits used in this job (pre-0.20).
This delays the job failure by a lot and hence degrades cluster utilization as a whole. If
i'm on a shared cluster with many jobs waiting on it to fail, it's bad. 
> The JT should prefer a failed task a lot earlier than waiting for a data local TT to
come around asking. 

-- 
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