From: Rohit Seth <rohitseth@google.com>
To: Dave Hansen <haveblue@us.ibm.com>
Cc: Kirill Korotaev <dev@sw.ru>,
"Martin J. Bligh" <mbligh@mbligh.org>,
vatsa@in.ibm.com, Alan Cox <alan@lxorguk.ukuu.org.uk>,
Andrew Morton <akpm@osdl.org>,
mingo@elte.hu, nickpiggin@yahoo.com.au, sam@vilain.net,
linux-kernel@vger.kernel.org, dev@openvz.org, efault@gmx.de,
balbir@in.ibm.com, sekharan@us.ibm.com, nagar@watson.ibm.com,
pj@sgi.com, Andrey Savochkin <saw@sw.ru>
Subject: Re: [RFC, PATCH 0/5] Going forward with Resource Management - A cpu controller
Date: Mon, 07 Aug 2006 12:00:57 -0700 [thread overview]
Message-ID: <1154977257.31962.57.camel@galaxy.corp.google.com> (raw)
In-Reply-To: <1154976236.19249.9.camel@localhost.localdomain>
On Mon, 2006-08-07 at 11:43 -0700, Dave Hansen wrote:
> On Mon, 2006-08-07 at 11:31 -0700, Rohit Seth wrote:
> > I think it is not a problem for OpenVZ because there is not that much
> > of
> > sharing going between containers as you mentioned (btw, this least
> > amount of sharing is a very good thing). Though I'm not sure if one
> > has
> > to go to the extent of doing fractions with memory accounting. If the
> > containers are set up in such a way that there is some sharing across
> > containers then it is okay to be unfair and charge one of those
> > containers for the specific resource completely.
>
> Right, and if you do reclaim against containers which are over their
> limits, the containers being unfairly charged will tend to get hit
> first. But, once this happens, I would hope that the ownership of those
> shared pages should settle out among all of the users.
>
I think there is lot of simplicity and value add by charging one
container (even unfairly) for one resource completely. This puts the
onus on system admin to set up the containers appropriately.
> If you have 100 containers sharing 100 pages, container0 might be
> charged for all 100 pages at first, but I'd hope that eventually
> containers 0->99 would each get charged for a single page.
>
You would be better off having a notion of "shared" container where
these kind of resources get charged. So, if 100 processes are all
touching pages from some common file then the file could be part of
self-contained container with its own limits etc.
-rohit
next prev parent reply other threads:[~2006-08-07 19:03 UTC|newest]
Thread overview: 78+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-08-04 5:07 Srivatsa Vaddagiri
2006-08-04 5:09 ` [ RFC, PATCH 1/5 ] CPU controller - base changes Srivatsa Vaddagiri
2006-08-04 7:35 ` Andrew Morton
2006-08-04 11:18 ` Srivatsa Vaddagiri
2006-08-04 14:34 ` Kirill Korotaev
2006-08-04 14:50 ` Balbir Singh
2006-08-04 14:51 ` Srivatsa Vaddagiri
2006-08-04 5:10 ` [ RFC, PATCH 2/5 ] CPU controller - Define group operations Srivatsa Vaddagiri
2006-08-04 23:10 ` Jiri Slaby
2006-08-04 5:11 ` [ RFC, PATCH 3/5 ] CPU controller - deal with movement of tasks Srivatsa Vaddagiri
2006-08-04 5:12 ` [ RFC, PATCH 4/5 ] CPU controller - deal with dont care groups Srivatsa Vaddagiri
2006-08-04 5:13 ` [ RFC, PATCH 5/5 ] CPU controller - interface with cpusets Srivatsa Vaddagiri
2006-08-04 5:36 ` [RFC, PATCH 0/5] Going forward with Resource Management - A cpu controller Andrew Morton
2006-08-04 5:42 ` Andrew Morton
2006-08-04 9:49 ` Alan Cox
2006-08-04 11:41 ` Srivatsa Vaddagiri
2006-08-04 14:51 ` Kirill Korotaev
2006-08-04 15:31 ` Srivatsa Vaddagiri
2006-08-04 16:03 ` Kirill Korotaev
2006-08-04 17:02 ` [ProbableSpam] " Shailabh Nagar
2006-08-04 18:27 ` Rohit Seth
2006-08-04 19:11 ` Shailabh Nagar
2006-08-04 19:24 ` Rohit Seth
2006-08-07 7:19 ` Kirill Korotaev
2006-08-07 17:14 ` Rohit Seth
2006-08-08 7:17 ` Kirill Korotaev
2006-08-08 17:16 ` Rohit Seth
2006-08-04 17:50 ` Martin Bligh
2006-08-07 7:25 ` Kirill Korotaev
2006-08-07 14:34 ` Martin J. Bligh
2006-08-07 16:33 ` Kirill Korotaev
2006-08-07 18:31 ` Rohit Seth
2006-08-07 18:43 ` Dave Hansen
2006-08-07 19:00 ` Rohit Seth [this message]
2006-08-07 19:46 ` Martin Bligh
2006-08-08 14:19 ` memory resource accounting (was Re: [RFC, PATCH 0/5] Going forward with Resource Management - A cpu controller) Nick Piggin
2006-08-08 14:57 ` Dave Hansen
2006-08-08 15:22 ` Nick Piggin
2006-08-09 13:43 ` Kirill Korotaev
2006-08-08 17:08 ` Martin Bligh
2006-08-09 1:54 ` Nick Piggin
2006-08-08 17:34 ` Rohit Seth
2006-08-09 4:33 ` Andi Kleen
2006-08-09 6:00 ` Magnus Damm
2006-08-09 6:06 ` Andi Kleen
2006-08-09 6:56 ` Andrey Savochkin
2006-08-08 7:19 ` [RFC, PATCH 0/5] Going forward with Resource Management - A cpu controller Kirill Korotaev
2006-08-04 16:16 ` Kirill Korotaev
2006-08-04 16:49 ` [ProbableSpam] " Shailabh Nagar
2006-08-04 17:03 ` Dipankar Sarma
2006-08-04 18:17 ` Shailabh Nagar
2006-08-07 7:23 ` Kirill Korotaev
2006-08-04 14:57 ` Kirill Korotaev
2006-08-04 5:58 ` Paul Jackson
2006-08-04 6:02 ` Paul Jackson
2006-08-04 6:16 ` Paul Jackson
2006-08-04 6:20 ` Dipankar Sarma
2006-08-04 6:31 ` Paul Jackson
2006-08-04 6:37 ` Dipankar Sarma
2006-08-04 6:49 ` Andrew Morton
2006-08-04 6:45 ` Andrew Morton
2006-08-04 7:10 ` Dipankar Sarma
2006-08-04 7:24 ` Andrew Morton
2006-08-04 19:10 ` Chandra Seetharaman
2006-08-04 6:56 ` Srivatsa Vaddagiri
2006-08-04 7:13 ` Andrew Morton
2006-08-04 11:16 ` Srivatsa Vaddagiri
2006-08-04 18:51 ` Andrew Morton
2006-08-04 14:20 ` Kirill Korotaev
2006-08-04 14:35 ` Christoph Hellwig
2006-08-04 15:29 ` [ProbableSpam] " Shailabh Nagar
2006-08-07 7:29 ` Kirill Korotaev
2006-08-07 9:30 ` Paul Jackson
2006-08-07 15:58 ` Chandra Seetharaman
2006-08-07 16:10 ` Kirill Korotaev
2006-08-07 17:15 ` Paul Jackson
2006-08-07 18:19 ` Rohit Seth
2006-08-05 3:30 ` Nick Piggin
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=1154977257.31962.57.camel@galaxy.corp.google.com \
--to=rohitseth@google.com \
--cc=akpm@osdl.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=balbir@in.ibm.com \
--cc=dev@openvz.org \
--cc=dev@sw.ru \
--cc=efault@gmx.de \
--cc=haveblue@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mbligh@mbligh.org \
--cc=mingo@elte.hu \
--cc=nagar@watson.ibm.com \
--cc=nickpiggin@yahoo.com.au \
--cc=pj@sgi.com \
--cc=sam@vilain.net \
--cc=saw@sw.ru \
--cc=sekharan@us.ibm.com \
--cc=vatsa@in.ibm.com \
/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
Powered by JetHome