From: Orit Wasserman <owasserm@redhat.com>
To: Peter Lieven <pl@dlhnet.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] [PATCH] migration: use XBZRLE only after bulk stage
Date: Mon, 04 Mar 2013 20:51:08 +0200 [thread overview]
Message-ID: <5134ED1C.5030202@redhat.com> (raw)
In-Reply-To: <54451010-DFA7-416A-9B8A-0762FB71DF36@dlhnet.de>
On 03/04/2013 07:10 PM, Peter Lieven wrote:
>
> Am 01.03.2013 um 17:04 schrieb Eric Blake <eblake@redhat.com>:
>
>> On 03/01/2013 07:50 AM, Peter Lieven wrote:
>>
>>> just a quick test on my desktop:
>>>
>>> ~/git/qemu$ x86_64-softmmu/qemu-system-x86_64 -k de -enable-kvm -m 1024
>>> -drive
>>> if=virtio,file=iscsi://172.21.200.31/iqn.2001-05.com.equallogic:0-8a0906-16470e107-713001aa6de511e0-001-test/0
>>> -vnc :1 -boot dc -monitor stdio
>>>
>>> using ubuntu 12.04.1 desktop and the example from docs/xbzrle.txt
>>
>> Thanks. Reformatting a bit:
>>
>>>
>>> a) with the patch
>>
>> designated with '+'
>>
>>> b) without the patch
>>
>> designated with '-'
>>
>> + total time: 22185 milliseconds
>> - total time: 22410 milliseconds
>>
>> Shaved 0.3 seconds, better than 1%!
>>
>> + downtime: 29 milliseconds
>> - downtime: 21 milliseconds
>>
>> Not sure why downtime seemed worse, but probably not the end of the world.
>>
>> + transferred ram: 706034 kbytes
>> - transferred ram: 721318 kbytes
>>
>> Fewer bytes sent - good.
>>
>> + remaining ram: 0 kbytes
>> - remaining ram: 0 kbytes
>> + total ram: 1057216 kbytes
>> - total ram: 1057216 kbytes
>> + duplicate: 108556 pages
>> - duplicate: 105553 pages
>> + normal: 175146 pages
>> - normal: 179589 pages
>> + normal bytes: 700584 kbytes
>> - normal bytes: 718356 kbytes
>>
>> Fewer normal bytes...
>>
>> + cache size: 67108864 bytes
>> - cache size: 67108864 bytes
>> + xbzrle transferred: 3127 kbytes
>> - xbzrle transferred: 630 kbytes
>>
>> ...and more compressed pages sent - good.
>>
>> + xbzrle pages: 117811 pages
>> - xbzrle pages: 21527 pages
>> + xbzrle cache miss: 18750
>> - xbzrle cache miss: 179589
>>
>> And very good improvement on the cache miss rate.
>>
>> + xbzrle overflow : 0
>> - xbzrle overflow : 0
>>
>> Thanks, this proves it's a good patch.
>
> At least for the artificially generated load.
>
> @Paolo: Have you seen my other question? Can the same page be transferred in the same round
> more than once? If yes, I have to improve the patch for that case.
The same page can't be transferred more than once in the same round.
Orit
>
> Peter
>
prev parent reply other threads:[~2013-03-04 18:49 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-01 13:31 [Qemu-devel] [PATCH] migration: use XBZRLE only after bulk stage Peter Lieven
2013-03-01 13:52 ` Eric Blake
2013-03-01 13:53 ` Paolo Bonzini
2013-03-01 14:06 ` Peter Lieven
2013-03-01 14:08 ` Eric Blake
2013-03-01 14:13 ` Peter Lieven
2013-03-01 14:23 ` Eric Blake
2013-03-01 14:50 ` Peter Lieven
2013-03-01 16:04 ` Eric Blake
2013-03-04 17:10 ` Peter Lieven
2013-03-04 17:33 ` Paolo Bonzini
2013-03-04 17:39 ` Peter Lieven
2013-03-04 18:51 ` Orit Wasserman [this message]
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=5134ED1C.5030202@redhat.com \
--to=owasserm@redhat.com \
--cc=pbonzini@redhat.com \
--cc=pl@dlhnet.de \
--cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).