[
https://issues.apache.org/jira/browse/YARN-3453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14573754#comment-14573754
]
Karthik Kambatla commented on YARN-3453:
----------------------------------------
Few comments:
# New imports in FairScheduler and FSLeafQueue are not required.
# Looking at the remaining uses of DefaultResourceCalculator in FairScheduler,
we could benefit from updating all of them to DominantResourceCalculator?
[~ashwinshankar77] - do you concur?
# In FairScheduler, changing the scope of RESOURCE_CALCULATOR and
DOMINANT_RESOURCE_CALCULATOR is not required.
# We should add unit-tests to avoid regressions in the future.
# Nit: In each of the policies, my preference would be not make the calculator
and comparator members static unless required. We have had cases where our
tests would invoke multiple instances of the class leading to issues. Not that
I foresee multiple instantiations for these classes, but would like to avoid it
if we can.
> Fair Scheduler : Parts of preemption logic uses DefaultResourceCalculator
> even in DRF mode causing thrashing
> ------------------------------------------------------------------------------------------------------------
>
> Key: YARN-3453
> URL: https://issues.apache.org/jira/browse/YARN-3453
> Project: Hadoop YARN
> Issue Type: Bug
> Components: fairscheduler
> Affects Versions: 2.6.0
> Reporter: Ashwin Shankar
> Assignee: Arun Suresh
> Attachments: YARN-3453.1.patch, YARN-3453.2.patch
>
>
> There are two places in preemption code flow where DefaultResourceCalculator
> is used, even in DRF mode.
> Which basically results in more resources getting preempted than needed, and
> those extra preempted containers aren’t even getting to the “starved” queue
> since scheduling logic is based on DRF's Calculator.
> Following are the two places :
> 1. {code:title=FSLeafQueue.java|borderStyle=solid}
> private boolean isStarved(Resource share)
> {code}
> A queue shouldn’t be marked as “starved” if the dominant resource usage
> is >= fair/minshare.
> 2. {code:title=FairScheduler.java|borderStyle=solid}
> protected Resource resToPreempt(FSLeafQueue sched, long curTime)
> {code}
> --------------------------------------------------------------
> One more thing that I believe needs to change in DRF mode is : during a
> preemption round,if preempting a few containers results in satisfying needs
> of a resource type, then we should exit that preemption round, since the
> containers that we just preempted should bring the dominant resource usage to
> min/fair share.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
