[
https://issues.apache.org/jira/browse/MESOS-3283?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Marco Massenzio reassigned MESOS-3283:
--------------------------------------
Assignee: Marco Massenzio
This correlates with some of our own observations and some customer reports.
Some work has already been done in this area, but it seems that it mostly
depends on Slaves being "well-behaved" in declining offers (indefinitely) when
they no longer need them (potentially, "reviving" when they need them again).
Assigning to myself for further triage (possibly also related to other work
already being done).
Thanks for reporting!
> Improve batch allocations performance especially with large number of slaves
> and frameworks.
> --------------------------------------------------------------------------------------------
>
> Key: MESOS-3283
> URL: https://issues.apache.org/jira/browse/MESOS-3283
> Project: Mesos
> Issue Type: Improvement
> Components: allocation
> Affects Versions: 0.23.0
> Reporter: Mandeep Chadha
> Assignee: Marco Massenzio
> Labels: mesosphere, tech-debt
>
> Improve batch allocations performance especially with large number of slaves
> and frameworks.
> e.g. these are the allocation timings for 10K slaves and varying number of
> frameworks.
> Using 10000 slaves and 1 frameworks
> Added 10000 slaves in 14.50836112secs
> Updated 10000 slaves in 18.665093703secs
> [ OK ]
> SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddAndUpdateSlave/12 (34983
> ms)
> [ RUN ]
> SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddAndUpdateSlave/13
> Using 10000 slaves and 50 frameworks
> Added 10000 slaves in 51.534229549secs
> Updated 10000 slaves in 57.131554303secs
> [ OK ]
> SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddAndUpdateSlave/13 (110449
> ms)
> [ RUN ]
> SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddAndUpdateSlave/14
> Using 10000 slaves and 100 frameworks
> Added 10000 slaves in 1.5891310434mins
> Updated 10000 slaves in 1.80562078148333mins
> [ OK ]
> SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddAndUpdateSlave/14 (205467
> ms)
> [ RUN ]
> SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddAndUpdateSlave/15
> Using 10000 slaves and 200 frameworks
> Added 10000 slaves in 3.0750647275mins
> Updated 10000 slaves in 3.85846762096667mins
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
