[
https://issues.apache.org/jira/browse/MAPREDUCE-5844?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14001941#comment-14001941
]
Jason Lowe commented on MAPREDUCE-5844:
---------------------------------------
Yes, the always-full-but-churning queue scenario can lead to an unfortunate
preemption case, because the RM and AM can't know cluster resources will free
up imminently. I think it's reasonable to add a configurable delay before
preempting a reducer when a map retroactively fails. That way users can tune
based on their cluster usage -- if they have a heavily used, high-churn cluster
they can tune the parameter based on the average time for a container to free
up in their queue. Other users who rarely run in this mode or for which the
delay would be intolerable can set this to zero to avoid unnecessary delays in
processing fetch-failed maps when there is no current headroom.
> Reducer Preemption is too aggressive
> ------------------------------------
>
> Key: MAPREDUCE-5844
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-5844
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Reporter: Maysam Yabandeh
> Assignee: Maysam Yabandeh
>
> We observed cases where the reducer preemption makes the job finish much
> later, and the preemption does not seem to be necessary since after
> preemption both the preempted reducer and the mapper are assigned
> immediately--meaning that there was already enough space for the mapper.
> The logic for triggering preemption is at
> RMContainerAllocator::preemptReducesIfNeeded
> The preemption is triggered if the following is true:
> {code}
> headroom + am * |m| + pr * |r| < mapResourceRequest
> {code}
> where am: number of assigned mappers, |m| is mapper size, pr is number of
> reducers being preempted, and |r| is the reducer size.
> The original idea apparently was that if headroom is not big enough for the
> new mapper requests, reducers should be preempted. This would work if the job
> is alone in the cluster. Once we have queues, the headroom calculation
> becomes more complicated and it would require a separate headroom calculation
> per queue/job.
> So, as a result headroom variable is kind of given up currently: *headroom is
> always set to 0* What this implies to the speculation is that speculation
> becomes very aggressive, not considering whether there is enough space for
> the mappers or not.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
