mesos-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benjamin Mahler <bmah...@apache.org>
Subject Re: GPU Users -- Deprecation of GPU_RESOURCES capability
Date Fri, 26 May 2017 20:48:46 GMT
I filed https://issues.apache.org/jira/browse/MESOS-7574 for reservations
to multiple roles. We'll find one that captures the deprecation of the
GPU_RESOURCES capability as well, with reservations to multiple roles as a
blocker.

On Fri, May 26, 2017 at 8:54 AM, Zhitao Li <zhitaoli.cs@gmail.com> wrote:

> Hi Benjamin,
>
> Thanks for getting back. Do you have an issue already filed for
> the "reservations to multiple roles" story, or is it folded under another
> JIRA story?
>
>
>
> On Fri, May 26, 2017 at 12:44 AM, Benjamin Mahler <bmahler@apache.org>
> wrote:
>
> > Thanks for the feedback!
> >
> > There have been some discussions for allowing reservations to multiple
> > roles (or more generally, role expressions), which is essentially what
> > you've suggested Zhitao. (However, note that what is provided by the GPU
> > capability filtering is not quite this, it's actually analogous to a
> > reservation for multiple schedulers, not roles). Reservations to multiple
> > roles seems to be the right replacement for those who rely on the GPU
> > filtering behavior.
> >
> > Since we don't have reservations to multiple roles at this point, we
> > shouldn't deprecate the GPU_RESOURCES capability until this is in place.
> >
> > With hierarchical roles, it's possible (although potentially clumsy) to
> > achieve roughly what is provided by the GPU filtering using sub-roles.
> > Since reservations made to a "gpu" role would be available to all of the
> > descendant roles within tree, e.g.
> > "gpu/analytics", "gpu/forecasting/training", etc. This is equivalent to a
> > restricted version of reservations to multiple roles, where the roles are
> > restricted to the descendant roles. This can get clumsy because if
> > "eng/backend/image-processing" wants to get in on the reserved gpus, the
> > user would have to place a related role underneath the "gpu" role, e.g.
> > "gpu/eng/backend/image-processing".
> >
>
> The exact reason you mentioned about the "clumsy" part would effectively
> prevent me of implementing this in our org even if it's already available.
>
>
> >
> > For the addition of the filter, note that this flag would be a temporary
> > measure that would be removed when the deprecation cycle of the
> capability
> > is complete. It would be good to independently consider the generalized
> > filtering idea you brought up.
> >
> > On Mon, May 22, 2017 at 9:15 AM, Zhitao Li <zhitaoli.cs@gmail.com>
> wrote:
> >
> > > Hi Kevin,
> > >
> > > Thanks for engaging with the community on this. My 2 cents:
> > >
> > > 1. I feel that this capabilities has a particular useful semantic which
> > is
> > > lacking in the current reservation system: reserving some scarce
> resource
> > > for a* dynamic list of multiple roles:*
> > >
> > > Right now, any reservation (static or dynamic) can only express the
> > > semantic of "reserving this resource for the given role R". However,
> in a
> > > complex cluster, it is possible that we have [R1, R2, ..., RN] which
> > wants
> > > to share the scarce resource among them but there is another set of
> roles
> > > which should never see the given resource.
> > >
> > > The new hierarchical role (and/or multi-role?) might be able to
> provide a
> > > better solution, but until that's widely available and adopted, the
> > > capabilities based hack is the only thing I know that can solve the
> > > problem.
> > >
> > > In fact, I think if we are going to wo with `--filter-gpu-resources`
> > path,
> > > I think we should make the filter more powerful (i.e, able to handle
> all
> > > known framework <-> resource/host constraints and more types of scarce
> > > resources) instead of the piecewise patches on a specific use case.
> > >
> > > Happy to chat more on this topic.
> > >
> > > On Sat, May 20, 2017 at 6:45 PM, Kevin Klues <klueska@gmail.com>
> wrote:
> > >
> > > > Hello GPU users,
> > > >
> > > > We are currently considering deprecating the requirement that
> > frameworks
> > > > register with the GPU _RESOURCES capability in order to receive
> offers
> > > that
> > > > contain GPUs. Going forward, we will recommend that users rely on
> > Mesos's
> > > > builtin `reservation` mechanism to achieve similar results.
> > > >
> > > > Before deprecating it, we wanted to get a sense from the community if
> > > > anyone is currently relying on this capability and would like to see
> it
> > > > persist. If not, we will begin deprecating it in the next Mesos
> release
> > > and
> > > > completely remove it in Mesos 2.0.
> > > >
> > > > As background, the original motivation for this capability was to
> keep
> > > > “legacy” frameworks from inadvertently scheduling jobs that don’t
> > require
> > > > GPUs on GPU capable machines and thus starving out other frameworks
> > that
> > > > legitimately want to place GPU jobs on those machines. The assumption
> > > here
> > > > was that most machines in a cluster won't have GPUs installed on
> them,
> > so
> > > > some mechanism was necessary to keep legacy frameworks from
> scheduling
> > > jobs
> > > > on those machines. In essence, it provided an implicit reservation of
> > GPU
> > > > machines for "GPU aware" frameworks, bypassing the traditional
> > > > `reservation` mechanism already built into Mesos.
> > > >
> > > > In such a setup, legacy frameworks would be free to schedule jobs on
> > > > non-GPU machines, and "GPU aware" frameworks would be free to
> schedule
> > > GPU
> > > > jobs GPU machines and other types of jobs on other machines (or mix
> and
> > > > match them however they please).
> > > >
> > > > However, the problem comes when *all* machines in a cluster contain
> > GPUs
> > > > (or even if most of the machines in a cluster container them). When
> > this
> > > is
> > > > the case, we have the opposite problem we were trying to solve by
> > > > introducing the GPU_RESOURCES capability in the first place. We end
> up
> > > > starving out jobs from legacy frameworks that *don’t* require GPU
> > > resources
> > > > because there are not enough machines available that don’t have GPUs
> on
> > > > them to service those jobs. We've actually seen this problem manifest
> > in
> > > > the wild at least once.
> > > >
> > > > An alternative to completely deprecating the GPU_RESOURCES flag would
> > be
> > > to
> > > > add a new flag to the mesos master called `--filter-gpu-resources`.
> > When
> > > > set to `true`, this flag will cause the mesos master to continue to
> > > > function as it does today. That is, it would filter offers containing
> > GPU
> > > > resources and only send them to frameworks that opt into the
> > > GPU_RESOURCES
> > > > framework capability. When set to `false`, this flag would cause the
> > > master
> > > > to *not* filter offers containing GPU resources, and indiscriminately
> > > send
> > > > them to all frameworks whether they set the GPU_RESOURCES capability
> or
> > > > not.
> > > >
> > > > , this flag would allow them to keep relying on it without
> disruption.
> > > >
> > > > We'd prefer to deprecate the capability completely, but would
> consider
> > > > adding this flag if people are currently relying on the GPU_RESOURCES
> > > > capability and would like to see it persist
> > > >
> > > > We welcome any feedback you have.
> > > >
> > > > Kevin + Ben
> > > >
> > >
> > >
> > >
> > > --
> > > Cheers,
> > >
> > > Zhitao Li
> > >
> >
>
>
>
> --
> Cheers,
>
> Zhitao Li
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message