qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Martin Mailand <martin@tuxadero.com>
To: Dongsu Park <dongsu.park@profitbricks.com>
Cc: stefanha@gmail.com, qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: [Qemu-devel] virtio-blk performance regression and qemu-kvm
Date: Mon, 05 Mar 2012 17:13:20 +0100	[thread overview]
Message-ID: <4F54E620.8060400@tuxadero.com> (raw)
In-Reply-To: <20120210143639.GA17883@gmail.com>

Am 10.02.2012 15:36, schrieb Dongsu Park:
> Recently I observed performance regression regarding virtio-blk,
> especially different IO bandwidths between qemu-kvm 0.14.1 and 1.0.
> So I want to share the benchmark results, and ask you what the reason
> would be.


Hi,
I think I found the problem, there is no regression in the code.
I think the problem is, that qmeu-kvm with the IO-Thread enabled doesn't 
produce enough cpu load to get the core to a higher cpu frequency, 
because the load is distributed to two threads.
If I change the cpu governor to "performance" the result from the master 
branch is better than from the v0.14.1 branch.
I get the same results on a serversystem without powermanagment activated.

@Dongsu Could you confirm those findings?


1. Test on i7 Laptop with Cpu governor "ondemand".

v0.14.1
bw=63492KB/s iops=15873
bw=63221KB/s iops=15805

v1.0
bw=36696KB/s iops=9173
bw=37404KB/s iops=9350

master
bw=36396KB/s iops=9099
bw=34182KB/s iops=8545

Change the Cpu governor to "performance"
master
bw=81756KB/s iops=20393
bw=81453KB/s iops=20257


2. Test on AMD Istanbul without powermanagement activated.

v0.14.1
bw=53167KB/s iops=13291
bw=61386KB/s iops=15346

v1.0
bw=43599KB/s iops=10899
bw=46288KB/s iops=11572

master
bw=60678KB/s iops=15169
bw=62733KB/s iops=15683

-martin

  parent reply	other threads:[~2012-03-05 16:13 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-02-10 14:36 [Qemu-devel] virtio-blk performance regression and qemu-kvm Dongsu Park
2012-02-12 23:55 ` Rusty Russell
2012-02-21 16:45   ` Dongsu Park
2012-02-21 22:16     ` Rusty Russell
2012-02-13 11:57 ` Stefan Hajnoczi
2012-02-21 15:57   ` Dongsu Park
2012-02-21 17:27     ` Stefan Hajnoczi
2012-02-22 16:48       ` Dongsu Park
2012-02-22 19:53         ` Stefan Hajnoczi
2012-02-28 16:39           ` Martin Mailand
2012-02-28 17:05             ` Stefan Hajnoczi
2012-02-28 17:15               ` Martin Mailand
2012-02-29  8:38                 ` Stefan Hajnoczi
2012-02-29 13:12                   ` Martin Mailand
2012-02-29 13:44                     ` Stefan Hajnoczi
2012-02-29 13:52                       ` Stefan Hajnoczi
2012-03-05 16:13 ` Martin Mailand [this message]
2012-03-05 16:35   ` Stefan Hajnoczi
2012-03-05 16:44     ` Martin Mailand
2012-03-06 12:59       ` Stefan Hajnoczi
2012-03-06 22:07         ` Reeted
2012-03-07  8:04           ` Stefan Hajnoczi
2012-03-07 14:21             ` Reeted
2012-03-07 14:33               ` Stefan Hajnoczi
2012-03-07 10:39         ` Martin Mailand
2012-03-07 11:21           ` Paolo Bonzini
2012-03-06 14:32   ` Dongsu Park

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=4F54E620.8060400@tuxadero.com \
    --to=martin@tuxadero.com \
    --cc=dongsu.park@profitbricks.com \
    --cc=kvm@vger.kernel.org \
    --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).