From: Avi Kivity <avi@redhat.com>
To: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Pavel Machek <pavel@ucw.cz>,
Dan Magenheimer <dan.magenheimer@oracle.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
jeremy@goop.org, hugh.dickins@tiscali.co.uk, ngupta@vflare.org,
JBeulich@novell.com, chris.mason@oracle.com,
kurt.hackel@oracle.com, dave.mccracken@oracle.com,
npiggin@suse.de, akpm@linux-foundation.org, riel@redhat.com
Subject: Re: Frontswap [PATCH 0/4] (was Transcendent Memory): overview
Date: Fri, 30 Apr 2010 10:13:40 +0300 [thread overview]
Message-ID: <4BDA8324.7090409@redhat.com> (raw)
In-Reply-To: <1272591924.23895.807.camel@nimitz>
On 04/30/2010 04:45 AM, Dave Hansen wrote:
>
> A large portion of CMM2's gain came from the fact that you could take
> memory away from guests without _them_ doing any work. If the system is
> experiencing a load spike, you increase load even more by making the
> guests swap. If you can just take some of their memory away, you can
> smooth that spike out. CMM2 and frontswap do that. The guests
> explicitly give up page contents that the hypervisor does not have to
> first consult with the guest before discarding.
>
Frontswap does not do this. Once a page has been frontswapped, the host
is committed to retaining it until the guest releases it. It's really
not very different from a synchronous swap device.
I think cleancache allows the hypervisor to drop pages without the
guest's immediate knowledge, but I'm not sure.
--
I have a truly marvellous patch that fixes the bug which this
signature is too narrow to contain.
WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi@redhat.com>
To: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Pavel Machek <pavel@ucw.cz>,
Dan Magenheimer <dan.magenheimer@oracle.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
jeremy@goop.org, hugh.dickins@tiscali.co.uk, ngupta@vflare.org,
JBeulich@novell.com, chris.mason@oracle.com,
kurt.hackel@oracle.com, dave.mccracken@oracle.com,
npiggin@suse.de, akpm@linux-foundation.org, riel@redhat.com
Subject: Re: Frontswap [PATCH 0/4] (was Transcendent Memory): overview
Date: Fri, 30 Apr 2010 10:13:40 +0300 [thread overview]
Message-ID: <4BDA8324.7090409@redhat.com> (raw)
In-Reply-To: <1272591924.23895.807.camel@nimitz>
On 04/30/2010 04:45 AM, Dave Hansen wrote:
>
> A large portion of CMM2's gain came from the fact that you could take
> memory away from guests without _them_ doing any work. If the system is
> experiencing a load spike, you increase load even more by making the
> guests swap. If you can just take some of their memory away, you can
> smooth that spike out. CMM2 and frontswap do that. The guests
> explicitly give up page contents that the hypervisor does not have to
> first consult with the guest before discarding.
>
Frontswap does not do this. Once a page has been frontswapped, the host
is committed to retaining it until the guest releases it. It's really
not very different from a synchronous swap device.
I think cleancache allows the hypervisor to drop pages without the
guest's immediate knowledge, but I'm not sure.
--
I have a truly marvellous patch that fixes the bug which this
signature is too narrow to contain.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2010-04-30 18:36 UTC|newest]
Thread overview: 163+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-04-22 13:42 Frontswap [PATCH 0/4] (was Transcendent Memory): overview Dan Magenheimer
2010-04-22 13:42 ` Dan Magenheimer
2010-04-22 15:28 ` Avi Kivity
2010-04-22 15:28 ` Avi Kivity
2010-04-22 15:48 ` Dan Magenheimer
2010-04-22 15:48 ` Dan Magenheimer
2010-04-22 16:13 ` Avi Kivity
2010-04-22 16:13 ` Avi Kivity
2010-04-22 20:15 ` Dan Magenheimer
2010-04-22 20:15 ` Dan Magenheimer
2010-04-23 9:48 ` Avi Kivity
2010-04-23 9:48 ` Avi Kivity
2010-04-23 13:47 ` Dan Magenheimer
2010-04-23 13:47 ` Dan Magenheimer
2010-04-23 13:57 ` Avi Kivity
2010-04-23 13:57 ` Avi Kivity
2010-04-23 14:43 ` Dan Magenheimer
2010-04-23 14:43 ` Dan Magenheimer
2010-04-23 14:52 ` Avi Kivity
2010-04-23 14:52 ` Avi Kivity
2010-04-23 15:00 ` Avi Kivity
2010-04-23 15:00 ` Avi Kivity
2010-04-23 16:26 ` Dan Magenheimer
2010-04-23 16:26 ` Dan Magenheimer
2010-04-24 18:25 ` Avi Kivity
2010-04-24 18:25 ` Avi Kivity
[not found] ` <1c02a94a-a6aa-4cbb-a2e6-9d4647760e91@default4BD43033.7090706@redhat.com>
2010-04-25 0:41 ` Dan Magenheimer
2010-04-25 0:41 ` Dan Magenheimer
2010-04-25 12:06 ` Avi Kivity
2010-04-25 12:06 ` Avi Kivity
2010-04-25 13:12 ` Dan Magenheimer
2010-04-25 13:12 ` Dan Magenheimer
2010-04-25 13:18 ` Avi Kivity
2010-04-25 13:18 ` Avi Kivity
2010-04-28 5:55 ` Pavel Machek
2010-04-28 5:55 ` Pavel Machek
2010-04-29 14:42 ` Dan Magenheimer
2010-04-29 14:42 ` Dan Magenheimer
2010-04-29 18:59 ` Avi Kivity
2010-04-29 18:59 ` Avi Kivity
2010-04-29 19:01 ` Avi Kivity
2010-04-29 19:01 ` Avi Kivity
2010-04-29 18:53 ` Avi Kivity
2010-04-29 18:53 ` Avi Kivity
2010-04-30 1:45 ` Dave Hansen
2010-04-30 1:45 ` Dave Hansen
2010-04-30 7:13 ` Avi Kivity [this message]
2010-04-30 7:13 ` Avi Kivity
2010-04-30 15:59 ` Dan Magenheimer
2010-04-30 15:59 ` Dan Magenheimer
2010-04-30 16:08 ` Dave Hansen
2010-04-30 16:08 ` Dave Hansen
2010-05-10 16:05 ` Martin Schwidefsky
2010-05-10 16:05 ` Martin Schwidefsky
2010-04-30 16:16 ` Avi Kivity
2010-04-30 16:16 ` Avi Kivity
[not found] ` <4BDB18CE.2090608@goop.org4BDB2069.4000507@redhat.com>
[not found] ` <3a62a058-7976-48d7-acd2-8c6a8312f10f@default20100502071059.GF1790@ucw.cz>
2010-04-30 16:43 ` Dan Magenheimer
2010-04-30 16:43 ` Dan Magenheimer
2010-04-30 17:10 ` Dave Hansen
2010-04-30 17:10 ` Dave Hansen
2010-04-30 18:08 ` Avi Kivity
2010-04-30 18:08 ` Avi Kivity
2010-04-30 17:52 ` Jeremy Fitzhardinge
2010-04-30 17:52 ` Jeremy Fitzhardinge
2010-04-30 18:24 ` Avi Kivity
2010-04-30 18:24 ` Avi Kivity
2010-04-30 18:59 ` Jeremy Fitzhardinge
2010-04-30 18:59 ` Jeremy Fitzhardinge
2010-05-01 8:28 ` Avi Kivity
2010-05-01 8:28 ` Avi Kivity
2010-05-01 17:10 ` Dan Magenheimer
2010-05-01 17:10 ` Dan Magenheimer
2010-05-02 7:11 ` Pavel Machek
2010-05-02 7:11 ` Pavel Machek
2010-05-02 15:05 ` Dan Magenheimer
2010-05-02 15:05 ` Dan Magenheimer
2010-05-02 20:06 ` Pavel Machek
2010-05-02 20:06 ` Pavel Machek
2010-05-02 21:05 ` Dan Magenheimer
2010-05-02 21:05 ` Dan Magenheimer
2010-05-02 7:57 ` Nitin Gupta
2010-05-02 7:57 ` Nitin Gupta
2010-05-02 16:06 ` Dan Magenheimer
2010-05-02 16:06 ` Dan Magenheimer
2010-05-02 16:48 ` Avi Kivity
2010-05-02 16:48 ` Avi Kivity
2010-05-02 17:22 ` Dan Magenheimer
2010-05-02 17:22 ` Dan Magenheimer
2010-05-03 9:39 ` Avi Kivity
2010-05-03 9:39 ` Avi Kivity
2010-05-03 14:59 ` Dan Magenheimer
2010-05-03 14:59 ` Dan Magenheimer
2010-05-02 15:35 ` Avi Kivity
2010-05-02 15:35 ` Avi Kivity
2010-05-02 17:06 ` Dan Magenheimer
2010-05-02 17:06 ` Dan Magenheimer
2010-05-03 8:46 ` Avi Kivity
2010-05-03 8:46 ` Avi Kivity
2010-05-03 16:01 ` Dan Magenheimer
2010-05-03 16:01 ` Dan Magenheimer
2010-05-03 19:32 ` Pavel Machek
2010-05-03 19:32 ` Pavel Machek
2010-04-30 16:04 ` Dave Hansen
2010-04-30 16:04 ` Dave Hansen
2010-04-23 15:56 ` Dan Magenheimer
2010-04-23 15:56 ` Dan Magenheimer
2010-04-24 18:22 ` Avi Kivity
2010-04-24 18:22 ` Avi Kivity
2010-04-25 0:30 ` Dan Magenheimer
2010-04-25 0:30 ` Dan Magenheimer
2010-04-25 12:11 ` Avi Kivity
2010-04-25 12:11 ` Avi Kivity
[not found] ` <c5062f3a-3232-4b21-b032-2ee1f2485ff0@default4BD44E74.2020506@redhat.com>
2010-04-25 13:37 ` Dan Magenheimer
2010-04-25 13:37 ` Dan Magenheimer
2010-04-25 14:15 ` Avi Kivity
2010-04-25 14:15 ` Avi Kivity
2010-04-25 15:29 ` Dan Magenheimer
2010-04-25 15:29 ` Dan Magenheimer
2010-04-26 6:01 ` Avi Kivity
2010-04-26 6:01 ` Avi Kivity
2010-04-26 12:45 ` Dan Magenheimer
2010-04-26 12:45 ` Dan Magenheimer
2010-04-26 13:48 ` Avi Kivity
2010-04-26 13:48 ` Avi Kivity
2010-04-27 12:56 ` Pavel Machek
2010-04-27 12:56 ` Pavel Machek
2010-04-27 14:32 ` Dan Magenheimer
2010-04-27 14:32 ` Dan Magenheimer
2010-04-29 13:02 ` Pavel Machek
2010-04-29 13:02 ` Pavel Machek
2010-04-27 11:52 ` Valdis.Kletnieks
2010-04-27 0:49 ` Jeremy Fitzhardinge
2010-04-27 0:49 ` Jeremy Fitzhardinge
2010-04-27 12:55 ` Pavel Machek
2010-04-27 12:55 ` Pavel Machek
2010-04-27 14:43 ` Nitin Gupta
2010-04-27 14:43 ` Nitin Gupta
2010-04-29 13:04 ` Pavel Machek
2010-04-29 13:04 ` Pavel Machek
2010-04-24 1:49 ` Nitin Gupta
2010-04-24 1:49 ` Nitin Gupta
2010-04-24 18:27 ` Avi Kivity
2010-04-24 18:27 ` Avi Kivity
2010-04-25 3:11 ` Nitin Gupta
2010-04-25 3:11 ` Nitin Gupta
2010-04-25 12:16 ` Avi Kivity
2010-04-25 12:16 ` Avi Kivity
2010-04-25 16:05 ` Nitin Gupta
2010-04-25 16:05 ` Nitin Gupta
2010-04-26 6:06 ` Avi Kivity
2010-04-26 6:06 ` Avi Kivity
2010-04-26 12:50 ` Dan Magenheimer
2010-04-26 12:50 ` Dan Magenheimer
2010-04-26 13:43 ` Avi Kivity
2010-04-26 13:43 ` Avi Kivity
2010-04-27 8:29 ` Dan Magenheimer
2010-04-27 8:29 ` Dan Magenheimer
2010-04-27 9:21 ` Avi Kivity
2010-04-27 9:21 ` Avi Kivity
2010-04-26 13:47 ` Nitin Gupta
2010-04-26 13:47 ` Nitin Gupta
2010-04-23 16:35 ` Jiahua
2010-04-23 16:35 ` Jiahua
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=4BDA8324.7090409@redhat.com \
--to=avi@redhat.com \
--cc=JBeulich@novell.com \
--cc=akpm@linux-foundation.org \
--cc=chris.mason@oracle.com \
--cc=dan.magenheimer@oracle.com \
--cc=dave.mccracken@oracle.com \
--cc=dave@linux.vnet.ibm.com \
--cc=hugh.dickins@tiscali.co.uk \
--cc=jeremy@goop.org \
--cc=kurt.hackel@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=ngupta@vflare.org \
--cc=npiggin@suse.de \
--cc=pavel@ucw.cz \
--cc=riel@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.