qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Jamie Lokier <jamie@shareable.org>
To: qemu-devel@nongnu.org
Cc: Chris Wright <chrisw@redhat.com>,
	Uri Lublin <uri.lublin@qumranet.com>,
	Anthony Liguori <aliguori@us.ibm.com>,
	kvm@vger.kernel.org
Subject: Re: [Qemu-devel] [PATCH 2/10] Allow the monitor to be suspended	during non-blocking op
Date: Thu, 11 Sep 2008 12:59:56 +0100	[thread overview]
Message-ID: <20080911115956.GB16427@shareable.org> (raw)
In-Reply-To: <48C8F00A.90900@qumranet.com>

Avi Kivity wrote:
> Jamie Lokier wrote:
> > Avi Kivity wrote:
> >   
> >> (logically we would copy all of the data of all block devices, but 
> >> that's not very practical, so we assume shared storage).
> >>     
> >
> > Speaking of that, if the guest RAM were a memory-mapped file, couldn't
> > that use shared storage too?
> 
> You would need a clustered filesystem that supports coherent mmap()s.

Yes, something like GFS.

> > You'd have to be careful: it would need a distributed filesystem with
> > coherent mappings (i.e. not NFS), but they do exist.
> >
> > I'm guessing that the bulk of time spent in migration/checkpointing is
> > saving the RAM image.  Using a memory-mapped file on shared storage
> > for RAM might make that faster.  (Or slower!).
> >   
> The memory needs to be transferred anyway, so total time would not
> change.  You could start running on the target sooner, though.

Actually the memory doesn't need to be tranferred.  Only actively used
pages need to be.  If you have an 8GB guest, 7.75GB of which is the
guest's filesystem cache from something you did earlier and is no
longer used, you just need to transfer 250MB and it can continue
running on the target.

-- Jamie

  reply	other threads:[~2008-09-11 12:00 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-09-09 19:49 [Qemu-devel] [PATCH 0/10] Live migration for QEMU Anthony Liguori
2008-09-09 19:49 ` [Qemu-devel] [PATCH 1/10] Refactor QEMUFile for live migration Anthony Liguori
2008-09-10 13:25   ` Chris Lalancette
2008-09-10 14:38   ` [Qemu-devel] " Glauber Costa
2008-09-10 15:05     ` Avi Kivity
2008-09-10 15:16     ` Anthony Liguori
2008-09-12 15:40   ` [Qemu-devel] " Blue Swirl
2008-09-09 19:49 ` [Qemu-devel] [PATCH 2/10] Allow the monitor to be suspended during non-blocking op Anthony Liguori
2008-09-10  6:52   ` Avi Kivity
2008-09-10 10:05     ` Daniel P. Berrange
2008-09-10 11:11       ` Avi Kivity
2008-09-10 11:14         ` Daniel P. Berrange
2008-09-10 15:36           ` Avi Kivity
2008-09-10 15:40             ` Anthony Liguori
2008-09-10 15:58         ` Jamie Lokier
2008-09-11 10:16           ` Avi Kivity
2008-09-11 11:59             ` Jamie Lokier [this message]
2008-09-10 13:07     ` Anthony Liguori
2008-09-10 13:26     ` Chris Lalancette
2008-09-10 10:01   ` Daniel P. Berrange
2008-09-10 13:11     ` Anthony Liguori
2008-09-09 19:49 ` [Qemu-devel] [PATCH 3/10] Add bdrv_flush_all() Anthony Liguori
2008-09-10 13:26   ` Chris Lalancette
2008-09-10 14:46     ` Glauber Costa
2008-09-10 15:19       ` Anthony Liguori
2008-09-10 15:32         ` Glauber Costa
2008-09-10 15:39         ` Avi Kivity
2008-09-10 16:37         ` Paul Brook
2008-09-12 15:43   ` Blue Swirl
2008-09-09 19:49 ` [Qemu-devel] [PATCH 4/10] Add dirty tracking for live migration Anthony Liguori
2008-09-10 14:52   ` Glauber Costa
2008-09-10 14:56     ` Anthony Liguori
2008-09-10 15:01       ` Glauber Costa
2008-09-09 19:49 ` [Qemu-devel] [PATCH 5/10] Add network announce function Anthony Liguori
2008-09-10 13:27   ` Chris Lalancette
2008-09-10 13:54     ` Anthony Liguori
2008-09-10 14:00     ` Avi Kivity
2008-09-09 19:49 ` [Qemu-devel] [PATCH 6/10] Introduce v3 of savevm protocol Anthony Liguori
2008-09-10  7:09   ` Avi Kivity
2008-09-09 19:49 ` [Qemu-devel] [PATCH 7/10] Switch the memory savevm handler to be "live" Anthony Liguori
2008-09-09 22:25   ` Jamie Lokier
2008-09-09 22:49     ` Anthony Liguori
2008-09-10  7:17   ` Avi Kivity
2008-09-10 13:10     ` Anthony Liguori
2008-09-09 19:50 ` [Qemu-devel] [PATCH 8/10] Introduce a buffered QEMUFile wrapper Anthony Liguori
2008-09-12 15:16   ` Blue Swirl
2008-09-09 19:50 ` [Qemu-devel] [PATCH 9/10] Introduce the UI components for live migration Anthony Liguori
2008-09-09 19:50 ` [Qemu-devel] [PATCH 10/10] TCP based " Anthony Liguori
2008-09-10 16:46   ` Blue Swirl
2008-09-10 16:51     ` Anthony Liguori
2008-09-11 12:13 ` [Qemu-devel] [PATCH 0/10] Live migration for QEMU Atsushi SAKAI
2008-09-11 13:06   ` Anthony Liguori
2008-09-11 13:30     ` Jamie Lokier
2008-09-11 14:12       ` Anthony Liguori
2008-09-11 15:32         ` Avi Kivity
2008-09-11 16:22           ` Anthony Liguori
2008-09-11 16:32             ` Avi Kivity

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=20080911115956.GB16427@shareable.org \
    --to=jamie@shareable.org \
    --cc=aliguori@us.ibm.com \
    --cc=chrisw@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=qemu-devel@nongnu.org \
    --cc=uri.lublin@qumranet.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;
as well as URLs for NNTP newsgroup(s).