From: Chandra Seetharaman <sekharan@us.ibm.com>
To: Kirill Korotaev <dev@sw.ru>
Cc: Dave Hansen <haveblue@us.ibm.com>, Rik van Riel <riel@redhat.com>,
CKRM-Tech <ckrm-tech@lists.sourceforge.net>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Andi Kleen <ak@suse.de>, Christoph Hellwig <hch@infradead.org>,
Andrey Savochkin <saw@sw.ru>,
devel@openvz.org, Hugh Dickins <hugh@veritas.com>,
Matt Helsley <matthltc@us.ibm.com>,
Alexey Dobriyan <adobriyan@mail.ru>,
Oleg Nesterov <oleg@tv-sign.ru>,
Alan Cox <alan@lxorguk.ukuu.org.uk>,
Pavel Emelianov <xemul@openvz.org>
Subject: Re: [ckrm-tech] [PATCH] BC: resource beancounters (v4) (added user memory)
Date: Wed, 06 Sep 2006 14:54:01 -0700 [thread overview]
Message-ID: <1157579641.31893.26.camel@linuxchandra> (raw)
In-Reply-To: <44FED3CA.7000005@sw.ru>
On Wed, 2006-09-06 at 17:57 +0400, Kirill Korotaev wrote:
> > On Tue, 2006-09-05 at 19:02 +0400, Kirill Korotaev wrote:
> >
> >>Core Resource Beancounters (BC) + kernel/user memory control.
> >>
> >>BC allows to account and control consumption
> >>of kernel resources used by group of processes.
> >
> >
> > Hi Kirill,
> >
> > I've honestly lost track of these discussions along the way, so I hope
> > you don't mind summarizing a bit.
> I think we need to create wiki to summarize it once and forever.
> http://wiki.openvz.org/UBC_discussion
>
> > Do these patches help with accounting for anything other than memory?
> this patch set - no, but the complete one - does:
> * numfile
> * numptys
> * numsocks (TCP, other, etc.)
> * numtasks
> * numflocks
> ...
> this list of resources was chosen to make sure that no DoS from the container
> is possible.
> This list is extensible easily and if resource is out of interest than
> its limits can be set to unlimited.
>
> > Will we need new user/kernel interfaces for cpu, i/o bandwidth, etc...?
> no. no new interfaces are required.
Good to know that.
Your CPU controller supports guarantee ?
Do you have a i/o controller ?
>
> BUT: I remind you the talks at OKS/OLS and in previous UBC discussions.
> It was noted that having a separate interfaces for CPU, I/O bandwidth
But, it will be lot simpler for the user to configure/use if they are
together. We should discuss this also.
> and memory maybe worthwhile. BTW, I/O bandwidth already has a separate
> interface :/
>
> > Have you given any thought to the possibility that a task might need to
> > move between accounting contexts? That has certainly been a
> > "requirement" pushed on to CKRM for a long time, and the need goes
> > something like this:
> Yes we thought about this and this is no more problematic for BC
> than for CKRM. See my explanation below.
>
> > 1. A system runs a web server, which services several virtual domains
> > 2. that web server receives a request for foo.com
> > 3. the web server switches into foo.com's accounting context
> > 4. the web server reads things from disk, allocates some memory, and
> > makes a database request.
> > 5. the database receives the request, and switches into foo.com's
> > accounting context, and charges foo.com for its resource use
> > etc...
> The question is - whether web server is multithreaded or not...
> If it is not - then no problem here, you can change current
> context and new resources will be charged accordingly.
>
> And current BC code is _able_ to handle it with _minor_ changes.
> (One just need to save bc not on mm struct, but rather on vma struct
> and change mm->bc on set_bc_id()).
>
> However, no one (can some one from CKRM team please?) explained so far
> what to do with threads. Consider the following example.
>
> 1. Threaded web server spawns a child to serve a client.
> 2. child thread touches some pages and they are charged to child BC
> (which differs from parent's one)
> 3. child exits, but since its mm is shared with parent, these pages
> stay mapped and charged to child BC.
>
> So the question is: what to do with these pages?
> - should we recharge them to another BC?
> - leave them charged?
Leave them charged. It will be charged to the appropriate UBC when they
touch it again.
>
> > So, the goal is to run _one_ copy of an application on a system, but
> > account for its resources in a much more fine-grained way than at the
> > application level.
> Yes.
>
> > I think we can probably use beancounters for this, if we do not worry
> > about migrating _existing_ charges when we change accounting context.
> > Does that make sense?
> exactly. thats what I'm saying. we can use beancounters for this
> if charges are kept for creator.
>
> Thanks,
> Kirill
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> ckrm-tech mailing list
> https://lists.sourceforge.net/lists/listinfo/ckrm-tech
--
----------------------------------------------------------------------
Chandra Seetharaman | Be careful what you choose....
- sekharan@us.ibm.com | .......you may get it.
----------------------------------------------------------------------
next prev parent reply other threads:[~2006-09-06 21:54 UTC|newest]
Thread overview: 144+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-05 15:02 [PATCH] BC: resource beancounters (v4) (added user memory) Kirill Korotaev
2006-09-05 15:19 ` [PATCH 1/13] BC: introduce atomic_dec_and_lock_irqsave() Kirill Korotaev
2006-09-05 15:20 ` [PATCH 2/13] BC: kconfig Kirill Korotaev
2006-09-05 15:21 ` [PATCH 3/17] BC: beancounters core (API) Kirill Korotaev
2006-09-05 15:23 ` [PATCH 4/13] BC: context inheriting and changing Kirill Korotaev
2006-09-05 15:24 ` [PATCH 5/13] BC: user interface (syscalls) Kirill Korotaev
2006-09-05 16:04 ` [ckrm-tech] " Balbir Singh
2006-09-06 8:29 ` Pavel Emelianov
2006-09-06 8:57 ` Balbir Singh
2006-09-06 10:42 ` Pavel Emelianov
2006-09-06 13:23 ` Balbir Singh
2006-09-06 13:45 ` Balbir Singh
2006-09-06 14:23 ` Kirill Korotaev
2006-09-05 15:25 ` [PATCH 6/13] BC: kernel memory (core) Kirill Korotaev
2006-09-05 15:26 ` [PATCH 7/13] BC: kernel memory (marks) Kirill Korotaev
2006-09-06 14:19 ` Cedric Le Goater
2006-09-05 15:27 ` [PATCH 8/13] BC: locked pages (core) Kirill Korotaev
2006-09-05 15:29 ` [PATCH 9/13] BC: locked pages (charge hooks) Kirill Korotaev
2006-09-06 3:43 ` Nick Piggin
2006-09-06 8:45 ` Pavel Emelianov
2006-09-06 9:41 ` Nick Piggin
2006-09-06 14:16 ` [ckrm-tech] " Kirill Korotaev
2006-09-05 15:30 ` [PATCH 10/13] BC: privvm pages Kirill Korotaev
2006-09-05 15:31 ` [PATCH 11/13] BC: vmrss (preparations) Kirill Korotaev
2006-09-05 22:09 ` Cedric Le Goater
2006-09-06 13:59 ` Kirill Korotaev
2006-09-07 16:28 ` [ckrm-tech] " Balbir Singh
2006-09-05 15:32 ` [PATCH 12/13] BC: vmrss (core) Kirill Korotaev
2006-09-05 15:33 ` [PATCH 13/13] BC: vmrss (charges) Kirill Korotaev
2006-09-05 16:53 ` [ckrm-tech] [PATCH] BC: resource beancounters (v4) (added user memory) Balbir Singh
2006-09-06 8:34 ` Pavel Emelianov
2006-09-06 13:06 ` Kirill Korotaev
2006-09-06 19:17 ` Balbir Singh
2006-09-06 22:06 ` Chandra Seetharaman
2006-09-07 3:08 ` Balbir Singh
2006-09-08 7:33 ` Pavel Emelianov
2006-09-08 15:43 ` Dave Hansen
2006-09-08 18:26 ` Balbir Singh
2006-09-11 6:56 ` Pavel Emelianov
2006-09-11 7:54 ` Balbir Singh
2006-09-11 8:13 ` Pavel Emelianov
2006-09-11 8:19 ` Balbir Singh
2006-09-12 10:39 ` Pavel Emelianov
2006-09-12 10:40 ` Pavel Emelianov
2006-09-12 12:01 ` Balbir Singh
2006-09-13 13:39 ` Pavel Emelianov
2006-09-11 10:21 ` Srivatsa Vaddagiri
2006-09-12 10:35 ` Pavel Emelianov
2006-09-11 18:49 ` Chandra Seetharaman
2006-09-12 10:48 ` Pavel Emelianov
2006-09-12 23:58 ` Chandra Seetharaman
2006-09-13 8:06 ` Pavel Emelianov
2006-09-13 12:15 ` Srivatsa Vaddagiri
2006-09-13 13:35 ` Pavel Emelianov
2006-09-13 22:31 ` Chandra Seetharaman
2006-09-14 7:53 ` Pavel Emelianov
2006-09-14 8:06 ` Balbir Singh
2006-09-14 13:02 ` Pavel Emelianov
2006-09-15 0:02 ` Chandra Seetharaman
2006-09-15 7:21 ` Pavel Emelianov
2006-09-15 8:49 ` Kirill Korotaev
2006-09-18 23:51 ` Chandra Seetharaman
2006-09-14 23:42 ` Chandra Seetharaman
2006-09-15 7:15 ` Pavel Emelianov
2006-09-15 8:55 ` Kirill Korotaev
2006-09-15 11:15 ` Pavel Emelianov
2006-09-18 8:25 ` Balbir Singh
2006-09-18 8:56 ` Pavel Emelianov
2006-09-18 11:20 ` Balbir Singh
2006-09-18 11:32 ` Pavel Emelianov
2006-09-19 0:05 ` Chandra Seetharaman
2006-09-19 8:04 ` Pavel Emelianov
2006-09-18 11:27 ` Balbir Singh
2006-09-18 12:37 ` Pavel Emelianov
2006-09-19 0:08 ` Chandra Seetharaman
2006-09-19 8:06 ` Pavel Emelianov
2006-09-18 23:48 ` Chandra Seetharaman
2006-09-11 18:44 ` Chandra Seetharaman
2006-09-15 8:57 ` Kirill Korotaev
2006-09-18 23:57 ` Chandra Seetharaman
2006-09-08 19:23 ` Chandra Seetharaman
2006-09-08 21:43 ` Rohit Seth
2006-09-11 18:25 ` Chandra Seetharaman
2006-09-11 19:10 ` Rohit Seth
2006-09-11 19:42 ` Chandra Seetharaman
2006-09-11 23:58 ` Rohit Seth
2006-09-12 9:53 ` Balbir Singh
2006-09-12 23:54 ` Chandra Seetharaman
2006-09-13 0:39 ` Rohit Seth
2006-09-13 1:10 ` Chandra Seetharaman
2006-09-13 1:25 ` Rohit Seth
2006-09-13 4:41 ` Srivatsa Vaddagiri
2006-09-13 22:20 ` Chandra Seetharaman
2006-09-14 1:22 ` Rohit Seth
2006-09-14 23:13 ` Chandra Seetharaman
2006-09-11 19:48 ` [Devel] " Kir Kolyshkin
2006-09-12 0:28 ` Rohit Seth
2006-09-12 10:44 ` Srivatsa Vaddagiri
2006-09-12 17:22 ` Rohit Seth
2006-09-12 17:40 ` Srivatsa Vaddagiri
2006-09-12 18:02 ` Rohit Seth
2006-09-13 0:02 ` Chandra Seetharaman
2006-09-13 0:43 ` Rohit Seth
2006-09-13 1:13 ` Chandra Seetharaman
2006-09-13 1:33 ` Rohit Seth
2006-09-13 22:24 ` Chandra Seetharaman
2006-09-14 1:27 ` Rohit Seth
2006-09-14 23:28 ` Chandra Seetharaman
2006-09-15 9:26 ` Kirill Korotaev
2006-09-15 16:52 ` Rohit Seth
2006-09-15 21:21 ` [Devel] " Kir Kolyshkin
2006-09-15 21:58 ` Rohit Seth
2006-09-19 0:02 ` [ckrm-tech] [Devel] " Chandra Seetharaman
2006-09-18 23:59 ` [ckrm-tech] " Chandra Seetharaman
2006-09-06 21:47 ` Chandra Seetharaman
2006-09-08 15:33 ` Dave Hansen
2006-09-08 15:57 ` [RFC] Add tgid aggregation to beancounters (was Re: [ckrm-tech] [PATCH] BC: resource beancounters (v4) (added user memory)) Balbir Singh
2006-09-11 21:24 ` V2: " Balbir Singh
2006-09-15 16:40 ` [ckrm-tech] V2: Add tgid aggregation to beancounters (was " Kirill Korotaev
2006-10-09 8:23 ` Balbir Singh
2006-09-05 17:46 ` [ckrm-tech] [PATCH] BC: resource beancounters (v4) (added user memory) Dave Hansen
2006-09-05 18:28 ` Balbir Singh
2006-09-06 0:17 ` Rohit Seth
2006-09-08 15:30 ` Dave Hansen
2006-09-08 17:10 ` Rohit Seth
2006-09-08 17:26 ` Shailabh Nagar
2006-09-08 21:15 ` Rohit Seth
2006-09-08 21:28 ` Shailabh Nagar
2006-09-06 13:57 ` Kirill Korotaev
2006-09-06 21:54 ` Chandra Seetharaman [this message]
2006-09-07 7:29 ` Pavel Emelianov
2006-09-07 19:16 ` Chandra Seetharaman
2006-09-08 7:22 ` Pavel Emelianov
2006-09-08 19:07 ` Chandra Seetharaman
2006-09-11 7:02 ` Pavel Emelianov
2006-09-11 13:04 ` Srivatsa Vaddagiri
2006-09-12 10:24 ` Pavel Emelianov
2006-09-12 10:29 ` Srivatsa Vaddagiri
2006-09-12 11:06 ` Pavel Emelianov
2006-09-12 12:04 ` Srivatsa Vaddagiri
2006-09-11 18:47 ` Chandra Seetharaman
2006-09-07 19:29 ` Chandra Seetharaman
2006-09-08 7:26 ` Pavel Emelianov
2006-09-08 19:10 ` Chandra Seetharaman
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=1157579641.31893.26.camel@linuxchandra \
--to=sekharan@us.ibm.com \
--cc=adobriyan@mail.ru \
--cc=ak@suse.de \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=ckrm-tech@lists.sourceforge.net \
--cc=dev@sw.ru \
--cc=devel@openvz.org \
--cc=haveblue@us.ibm.com \
--cc=hch@infradead.org \
--cc=hugh@veritas.com \
--cc=linux-kernel@vger.kernel.org \
--cc=matthltc@us.ibm.com \
--cc=oleg@tv-sign.ru \
--cc=riel@redhat.com \
--cc=saw@sw.ru \
--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