[
https://issues.apache.org/jira/browse/MAPREDUCE-1221?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12838237#action_12838237
]
dhruba borthakur commented on MAPREDUCE-1221:
---------------------------------------------
> If tasks, and hence jobs, are not penalized i.e. killed rather than failed,
> what is the incentive for the authors of the bad
>From my understanding, this JIRA is not about making users write good code. It
>is about making nodes in the cluster not die and rot (via excessive swapping,
>aka MAPREDUCE-257) when a huge-memory-consuming job comes along. It is also
>about letting other good jobs continue to run peacefully as much as possible.
>In a true utility model, the service provider provides service and it is upto
>the customer to write/deploy whenever code he wants to run on it (as long as
>he pays for it, of course:-))
> So killing some tasks and letting others proceed may still be better than not
> killing. That said, I agree that we can fine tune the policy
+1 for Hong's proposal. If somebody has a better policy of which tasks to kill
when this situation arises that will be great. The default policy is to kill
tasks with the maximum memory used and I think that is a very practical one,
isn't it?
> Kill tasks on a node if the free physical memory on that machine falls below
> a configured threshold
> ---------------------------------------------------------------------------------------------------
>
> Key: MAPREDUCE-1221
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-1221
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: tasktracker
> Affects Versions: 0.22.0
> Reporter: dhruba borthakur
> Assignee: Scott Chen
> Fix For: 0.22.0
>
> Attachments: MAPREDUCE-1221-v1.patch, MAPREDUCE-1221-v2.patch,
> MAPREDUCE-1221-v3.patch
>
>
> The TaskTracker currently supports killing tasks if the virtual memory of a
> task exceeds a set of configured thresholds. I would like to extend this
> feature to enable killing tasks if the physical memory used by that task
> exceeds a certain threshold.
> On a certain operating system (guess?), if user space processes start using
> lots of memory, the machine hangs and dies quickly. This means that we would
> like to prevent map-reduce jobs from triggering this condition. From my
> understanding, the killing-based-on-virtual-memory-limits (HADOOP-5883) were
> designed to address this problem. This works well when most map-reduce jobs
> are Java jobs and have well-defined -Xmx parameters that specify the max
> virtual memory for each task. On the other hand, if each task forks off
> mappers/reducers written in other languages (python/php, etc), the total
> virtual memory usage of the process-subtree varies greatly. In these cases,
> it is better to use kill-tasks-using-physical-memory-limits.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
