hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sunil G (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-4108) CapacityScheduler: Improve preemption to preempt only those containers that would satisfy the incoming request
Date Mon, 15 Feb 2016 17:16:18 GMT

    [ https://issues.apache.org/jira/browse/YARN-4108?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15147605#comment-15147605
] 

Sunil G commented on YARN-4108:
-------------------------------

Hi [~leftnoteasy]
Thanks for sharing full version of updated patch, really appreciate the efforts. I have few
dbts/comments in this patch. Pls help to check the same. :-)

1.
{{updateResToObtainAndKillableContainers}} tries to raise cancel preemption back to scheduler
if container resource is empty. So could we also remove from {{resToObtainByPartition}} here?
{code}
+        if (!partitionsHasResToObtain.contains(partition)) {
+          // When we don't need to preempt any resources from the queue/partition
+          // , cancel all killable containers from the queue
+          rmContext.getDispatcher().getEventHandler().handle(
+              new ContainerPreemptEvent(
+                  killableContainer.getApplicationAttemptId(),
+                  killableContainer,
+                  SchedulerEventType.MARK_CONTAINER_FOR_NONKILLABLE))
{code}

2. If {{conf.getLazyPreemptionEnabled()}} is disabled, do we need to handle events like MARK_CONTAINER_FOR_NONKILLABLE
in scheduler?

3. {{cleanupCompletedKillableContainers()}} in ProportionalCapacityPreemptionPolicy operates
on full live containers in cluster. And this can be a very big number, so we may hold the
concurrent map's access on each iteration of preemption policy. Will this be a pblm? Can we
optimize this further. For eg, could we set a boolean state within RMContainer to indicate
its not a live container?

4. {{killToPreemptContainers}} in LeafQueue invokes completedContainer on application's leafQueue.
I am thinking on some corner case where one {{LeafQueue#killToPreemptContainers}} invocation
results in a call to completedContainer of another leafQueue. Do you see some chance for same,
may be a moveQueue operation?


> CapacityScheduler: Improve preemption to preempt only those containers that would satisfy
the incoming request
> --------------------------------------------------------------------------------------------------------------
>
>                 Key: YARN-4108
>                 URL: https://issues.apache.org/jira/browse/YARN-4108
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: capacity scheduler
>            Reporter: Wangda Tan
>            Assignee: Wangda Tan
>         Attachments: YARN-4108-design-doc-V3.pdf, YARN-4108-design-doc-v1.pdf, YARN-4108-design-doc-v2.pdf,
YARN-4108.1.patch, YARN-4108.2.patch, YARN-4108.poc.1.patch, YARN-4108.poc.2-WIP.patch, YARN-4108.poc.3-WIP.patch,
YARN-4108.poc.4-WIP.patch
>
>
> This is sibling JIRA for YARN-2154. We should make sure container preemption is more
effective.
> *Requirements:*:
> 1) Can handle case of user-limit preemption
> 2) Can handle case of resource placement requirements, such as: hard-locality (I only
want to use rack-1) / node-constraints (YARN-3409) / black-list (I don't want to use rack1
and host\[1-3\])
> 3) Can handle preemption within a queue: cross user preemption (YARN-2113), cross applicaiton
preemption (such as priority-based (YARN-1963) / fairness-based (YARN-3319)).



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

Mime
View raw message