From: Balbir Singh <balbir@linux.vnet.ibm.com>
To: Paul Menage <menage@google.com>
Cc: bharata@linux.vnet.ibm.com, linux-kernel@vger.kernel.org,
Dhaval Giani <dhaval@linux.vnet.ibm.com>,
Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
Gautham R Shenoy <ego@in.ibm.com>,
Srivatsa Vaddagiri <vatsa@in.ibm.com>,
Ingo Molnar <mingo@elte.hu>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Pavel Emelyanov <xemul@openvz.org>, Avi Kivity <avi@redhat.com>,
kvm@vger.kernel.org,
Linux Containers <containers@lists.linux-foundation.org>,
Herbert Poetzl <herbert@13thfloor.at>
Subject: Re: [RFC] CPU hard limits
Date: Fri, 5 Jun 2009 17:55:27 +0800 [thread overview]
Message-ID: <20090605095527.GM11755@balbir.in.ibm.com> (raw)
In-Reply-To: <6599ad830906050248m2c569e5bx44fb3bbddf46f8b1@mail.gmail.com>
* menage@google.com <menage@google.com> [2009-06-05 02:48:36]:
> On Fri, Jun 5, 2009 at 2:36 AM, Balbir Singh<balbir@linux.vnet.ibm.com> wrote:
> >
> > The important scenario I have is adding and removing groups.
> >
> > Consider 10 cgroups with shares of 10 each, what if 5 new are created
> > with the same shares? We now start getting 100/15, even though we did
> > not change our shares.
>
> Are you assuming that arbitrary users can create new cgroups whenever
> they like, with whatever shares they like? In that situation, how
> would you use hard limits to provide guarantees? Presumably if the
> user could create a cgroup with an arbitrary share, they could create
> one with an arbitrary hard limit too.
>
What about applications running as root, that can create their own
groups? How about multiple instances of the same application started?
Do applications need to know that creating a group will hurt
guarantees provided to others?
> Can you explain a bit more about how you're envisaging cgroups being
> created, and how their shares and hard limits would get set? I was
> working on the assumption that (for any sub-tree of the CFS hierarchy)
> there's a single managing entity that gets to decide the shares given
> to the cgroups within that tree. That managing entity would be
> responsible for ensuring that the shares given out allowed guarantees
> to be met (or alternatively, that the probability of violating those
> guarantees based on the shares given out was within some tolerance
> threshold).
>
The point is that there is no single control entity for creating
groups. if run a solution, it might create groups without telling the
user. No one is arbitrating, not even libcgroup. What if someone
changes the cpuset assignment and moves CPUS x to y in an exclusive
cpuset all of a sudden. How do we arbitrate?
--
Balbir
next prev parent reply other threads:[~2009-06-05 9:56 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-04 5:36 Bharata B Rao
2009-06-04 12:19 ` Avi Kivity
2009-06-04 21:32 ` Mike Waychison
2009-06-05 3:03 ` Bharata B Rao
2009-06-05 3:33 ` Avi Kivity
2009-06-05 4:37 ` Balbir Singh
2009-06-05 4:44 ` Avi Kivity
2009-06-05 4:49 ` Balbir Singh
2009-06-05 5:09 ` Chris Friesen
2009-06-05 5:13 ` Balbir Singh
2009-06-05 5:10 ` Balbir Singh
2009-06-05 5:21 ` Avi Kivity
2009-06-05 5:27 ` Balbir Singh
2009-06-05 5:31 ` Bharata B Rao
2009-06-05 6:01 ` Avi Kivity
2009-06-05 8:16 ` Bharata B Rao
2009-06-07 6:04 ` Avi Kivity
2009-06-07 16:14 ` Bharata B Rao
2009-06-05 9:39 ` Balbir Singh
2009-06-05 13:14 ` Avi Kivity
2009-06-05 13:42 ` Balbir Singh
2009-06-07 6:09 ` Avi Kivity
2009-06-05 14:54 ` Chris Friesen
2009-06-07 6:10 ` Avi Kivity
2009-06-05 9:24 ` Balbir Singh
2009-06-05 6:03 ` Avi Kivity
2009-06-05 6:32 ` Bharata B Rao
2009-06-05 12:57 ` Avi Kivity
2009-06-05 5:16 ` Avi Kivity
2009-06-05 5:20 ` Balbir Singh
2009-06-05 3:07 ` Balbir Singh
2009-06-05 8:53 ` Paul Menage
2009-06-05 9:27 ` Bharata B Rao
2009-06-05 9:32 ` Paul Menage
2009-06-05 9:48 ` Dhaval Giani
2009-06-05 9:51 ` Paul Menage
2009-06-05 9:59 ` Dhaval Giani
2009-06-05 10:03 ` Paul Menage
2009-06-08 8:50 ` Pavel Emelyanov
2009-06-05 9:36 ` Balbir Singh
2009-06-05 9:48 ` Paul Menage
2009-06-05 9:55 ` Balbir Singh [this message]
2009-06-05 9:57 ` Paul Menage
2009-06-05 10:02 ` Paul Menage
2009-06-05 11:32 ` Srivatsa Vaddagiri
2009-06-05 12:18 ` Paul Menage
2009-06-07 10:11 ` Srivatsa Vaddagiri
2009-06-07 15:35 ` Balbir Singh
2009-06-08 4:37 ` Srivatsa Vaddagiri
2009-06-05 14:44 ` Chris Friesen
2009-06-05 13:02 ` Avi Kivity
2009-06-05 13:43 ` Dhaval Giani
2009-06-05 14:45 ` Chris Friesen
2009-06-05 9:02 ` Reinhard Tartler
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20090605095527.GM11755@balbir.in.ibm.com \
--to=balbir@linux.vnet.ibm.com \
--cc=a.p.zijlstra@chello.nl \
--cc=avi@redhat.com \
--cc=bharata@linux.vnet.ibm.com \
--cc=containers@lists.linux-foundation.org \
--cc=dhaval@linux.vnet.ibm.com \
--cc=ego@in.ibm.com \
--cc=herbert@13thfloor.at \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=menage@google.com \
--cc=mingo@elte.hu \
--cc=svaidy@linux.vnet.ibm.com \
--cc=vatsa@in.ibm.com \
--cc=xemul@openvz.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®