qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Avi Kivity <avi@redhat.com>
To: "Shribman, Aidan" <aidan.shribman@sap.com>
Cc: Blue Swirl <blauwirbel@gmail.com>,
	Stefan Hajnoczi <stefanha@gmail.com>,
	qemu-devel Developers <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] [PATCH v4] XBZRLE delta for live migration of large memory apps
Date: Mon, 08 Aug 2011 17:46:06 +0300	[thread overview]
Message-ID: <4E3FF6AE.8030004@redhat.com> (raw)
In-Reply-To: <AB5A8C7661872E428D6B8E1C2DFA35085D84AF0480@DEWDFECCR02.wdf.sap.corp>

On 08/08/2011 11:42 AM, Shribman, Aidan wrote:
> Subject: [PATCH v4] XBZRLE delta for live migration of large memory apps
> From: Aidan Shribman<aidan.shribman@sap.com>
>
> By using XBZRLE (Xor Binary Zero Run-Length-Encoding) we can reduce VM downtime
> and total live-migration time of VMs running memory write intensive workloads
> typical of large enterprise applications such as SAP ERP Systems, and generally
> speaking for any application with a sparse memory update pattern.
>
> On the sender side XBZRLE is used as a compact delta encoding of page updates,
> retrieving the old page content from an LRU cache (default size of 64 MB). The
> receiving side uses the existing page content and XBZRLE to decode the new page
> content.
>
> Work was originally based on research results published VEE 2011: Evaluation of
> Delta Compression Techniques for Efficient Live Migration of Large Virtual
> Machines by Benoit, Svard, Tordsson and Elmroth. Additionally the delta encoder
> XBRLE was improved further using XBZRLE instead.
>
> XBZRLE has a sustained bandwidth of 2-2.5 GB/s for typical workloads making it
> ideal for in-line, real-time encoding such as is needed for live-migration.
>
> A typical usage scenario:
>      {qemu} migrate_set_cachesize 256m
>      {qemu} migrate -x -d tcp:destination.host:4444
>      {qemu} info migrate
>      ...
>      transferred ram-duplicate: A kbytes
>      transferred ram-duplicate: B pages
>      transferred ram-normal: C kbytes
>      transferred ram-normal: D pages
>      transferred ram-xbrle: E kbytes
>      transferred ram-xbrle: F pages
>      overflow ram-xbrle: G pages
>      cache-hit ram-xbrle: H pages
>      cache-lookup ram-xbrle: J pages
>
> Testing: live migration with XBZRLE completed in 110 seconds, without live
> migration was not able to complete.
>
> A simple synthetic memory r/w load generator:
> ..    include<stdlib.h>
> ..    include<stdio.h>
> ..    int main()
> ..    {
> ..        char *buf = (char *) calloc(4096, 4096);
> ..        while (1) {
> ..            int i;
> ..            for (i = 0; i<  4096 * 4; i++) {
> ..                buf[i * 4096 / 4]++;
> ..            }
> ..            printf(".");
> ..        }
> ..    }
>
>


Please provide documentation in docs/ of the compression format.

IMO it should be disabled by default (with an option to disable it, via, 
sat, migrate-set-options, so we can migrate to older hosts).

The protocol should allow XBZRLE to turn itself off if it detects that 
it isn't effective.

-- 
error compiling committee.c: too many arguments to function

  parent reply	other threads:[~2011-08-08 14:46 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-08-08  8:42 [Qemu-devel] [PATCH v4] XBZRLE delta for live migration of large memory apps Shribman, Aidan
2011-08-08 13:29 ` Anthony Liguori
2011-08-08 13:41   ` Alexander Graf
2011-08-08 13:46     ` Anthony Liguori
2011-08-08 13:49     ` Avi Kivity
2011-08-08 13:51   ` Avi Kivity
2011-08-08 14:15     ` Anthony Liguori
2011-08-08 14:23       ` Avi Kivity
2011-08-08 14:33         ` Anthony Liguori
2011-08-08 14:39           ` Avi Kivity
2011-08-08 15:08             ` Anthony Liguori
2011-08-08 14:04   ` [Qemu-devel] [libvirt] " Daniel P. Berrange
2011-08-08 14:42     ` Avi Kivity
2011-08-08 14:46 ` Avi Kivity [this message]
2011-08-08 14:47   ` [Qemu-devel] " Avi Kivity
2011-08-08 14:56     ` Stefan Hajnoczi
2011-08-08 15:01       ` Avi Kivity
2011-08-08 15:10     ` Anthony Liguori
2011-08-08 15:15       ` Avi Kivity
2011-08-08 16:19         ` Anthony Liguori
2011-08-08 16:53           ` Avi Kivity
2011-08-08 16:55             ` Anthony Liguori
2011-08-10 15:07               ` Shribman, Aidan
2011-08-10 15:12                 ` Avi Kivity
2011-08-10 15:58                   ` Anthony Liguori
2011-08-10 16:08                     ` Avi Kivity
2011-08-10 16:23                       ` Anthony Liguori
2011-08-10 16:40                         ` Avi Kivity
2011-08-10 19:27                           ` Anthony Liguori
2011-08-11  8:03                             ` Shribman, Aidan
2011-08-11 13:00                               ` Anthony Liguori
2011-08-11  8:17                             ` Avi Kivity
2011-08-11  9:16                               ` [Qemu-devel] [libvirt] " Daniel P. Berrange
2011-08-11  9:20                                 ` Avi Kivity
2011-08-11 13:03                               ` [Qemu-devel] " Anthony Liguori
2011-08-11  9:24                             ` [Qemu-devel] [libvirt] " Daniel P. Berrange

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=4E3FF6AE.8030004@redhat.com \
    --to=avi@redhat.com \
    --cc=aidan.shribman@sap.com \
    --cc=blauwirbel@gmail.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@gmail.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).