[
https://issues.apache.org/jira/browse/HADOOP-2141?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12542895
]
Owen O'Malley commented on HADOOP-2141:
---------------------------------------
{quote}
There are always going to be such corner cases, and proper speculative
execution is the right solution to such problems.
{quote}
This is exactly backwards. Speculative execution absolutely can *not* be used
as a reliability solution. Applications can and do turn it off. Therefore, the
system must be completely reliable without speculative execution. Furthermore,
if there is a "freezing" problem it may well strike the speculative task also,
which would lock the job. Speculative execution is a pure optimization. It is
not for reliability.
> speculative execution start up condition based on completion time
> -----------------------------------------------------------------
>
> Key: HADOOP-2141
> URL: https://issues.apache.org/jira/browse/HADOOP-2141
> Project: Hadoop
> Issue Type: Improvement
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Koji Noguchi
> Assignee: Arun C Murthy
> Fix For: 0.16.0
>
>
> We had one job with speculative execution hang.
> 4 reduce tasks were stuck with 95% completion because of a bad disk.
> Devaraj pointed out
> bq . One of the conditions that must be met for launching a speculative
> instance of a task is that it must be at least 20% behind the average
> progress, and this is not true here.
> It would be nice if speculative execution also starts up when tasks stop
> making progress.
> Devaraj suggested
> bq. Maybe, we should introduce a condition for average completion time for
> tasks in the speculative execution check.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
