[jira] [Updated] (YARN-6406) Garbage Collect unused SchedulerRequestKeys
[
https://issues.apache.org/jira/browse/YARN-6406?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Arun Suresh updated YARN-6406:
--
Attachment: YARN-6406.002.patch
Updating patch to revert changes to {{AppInfo}} and update the
{{TestRMWebServicesApps}} test based on [~leftnoteasy]'s suggestion.
> Garbage Collect unused SchedulerRequestKeys
> ---
>
> Key: YARN-6406
> URL: https://issues.apache.org/jira/browse/YARN-6406
> Project: Hadoop YARN
> Issue Type: Improvement
>Affects Versions: 2.8.0, 2.7.3, 3.0.0-alpha2
>Reporter: Arun Suresh
>Assignee: Arun Suresh
> Attachments: YARN-6406.001.patch, YARN-6406.002.patch
>
>
> YARN-5540 introduced some optimizations to remove satisfied SchedulerKeys
> from the AppScheduleingInfo. It looks like after YARN-6040,
> ScedulerRequestKeys are removed only if the Application sends a 0
> numContainers requests. While earlier, the outstanding schedulerKeys were
> also remove as soon as a container is allocated as well.
> An additional optimization we were hoping to include is to remove the
> ResourceRequests itself once the numContainers == 0, since we see in our
> clusters that the RM heap space consumption increases drastically due to a
> large number of ResourceRequests with 0 num containers.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6406) Garbage Collect unused SchedulerRequestKeys
[ https://issues.apache.org/jira/browse/YARN-6406?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Arun Suresh updated YARN-6406: -- Affects Version/s: 2.8.0 2.7.3 3.0.0-alpha2 > Garbage Collect unused SchedulerRequestKeys > --- > > Key: YARN-6406 > URL: https://issues.apache.org/jira/browse/YARN-6406 > Project: Hadoop YARN > Issue Type: Improvement >Affects Versions: 2.8.0, 2.7.3, 3.0.0-alpha2 >Reporter: Arun Suresh >Assignee: Arun Suresh > Attachments: YARN-6406.001.patch > > > YARN-5540 introduced some optimizations to remove satisfied SchedulerKeys > from the AppScheduleingInfo. It looks like after YARN-6040, > ScedulerRequestKeys are removed only if the Application sends a 0 > numContainers requests. While earlier, the outstanding schedulerKeys were > also remove as soon as a container is allocated as well. > An additional optimization we were hoping to include is to remove the > ResourceRequests itself once the numContainers == 0, since we see in our > clusters that the RM heap space consumption increases drastically due to a > large number of ResourceRequests with 0 num containers. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6406) Garbage Collect unused SchedulerRequestKeys
[ https://issues.apache.org/jira/browse/YARN-6406?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Arun Suresh updated YARN-6406: -- Attachment: YARN-6406.001.patch Attaching initial patch. * Moved the schedulerKeys dataStructure to a ConcurrentSkipListSet * Removed ANY ResourceRequests once num container == 0 * Removed the SchedulerPlacementSet entry completely if no more resourceRequests exists * Added some tests. It looks like SchedulerKeys are actually removed once Containers are allocated. This patch additionally removes ResourceRequest objects which are SchedulerRequestKey key-ed entries as well. > Garbage Collect unused SchedulerRequestKeys > --- > > Key: YARN-6406 > URL: https://issues.apache.org/jira/browse/YARN-6406 > Project: Hadoop YARN > Issue Type: Improvement >Reporter: Arun Suresh >Assignee: Arun Suresh > Attachments: YARN-6406.001.patch > > > YARN-5540 introduced some optimizations to remove satisfied SchedulerKeys > from the AppScheduleingInfo. It looks like after YARN-6040, > ScedulerRequestKeys are removed only if the Application sends a 0 > numContainers requests. While earlier, the outstanding schedulerKeys were > also remove as soon as a container is allocated as well. > An additional optimization we were hoping to include is to remove the > ResourceRequests itself once the numContainers == 0, since we see in our > clusters that the RM heap space consumption increases drastically due to a > large number of ResourceRequests with 0 num containers. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
