From: Josef Bacik <josef@redhat.com>
To: Tomasz Chmielewski <mangoo@wpkg.org>
Cc: linux-kernel@vger.kernel.org, linux-btrfs@vger.kernel.org,
hch@infradead.org, gg.mariotti@gmail.com,
"Justin P. Mattock" <justinmattock@gmail.com>,
mjt@tls.msk.ru, josef@redhat.com, tytso@mit.edu
Subject: Re: BTRFS: Unbelievably slow with kvm/qemu
Date: Sun, 29 Aug 2010 20:14:41 -0400 [thread overview]
Message-ID: <20100830001441.GA838@dhcp231-156.rdu.redhat.com> (raw)
In-Reply-To: <4C7AB645.5090804@wpkg.org>
On Sun, Aug 29, 2010 at 09:34:29PM +0200, Tomasz Chmielewski wrote:
> Christoph Hellwig wrote:
>
>> There are a lot of variables when using qemu.
>>
>> The most important one are:
>>
>> - the cache mode on the device. The default is cache=writethrough,
>> which is not quite optimal. You generally do want to use cache=none
>> which uses O_DIRECT in qemu.
>> - if the backing image is sparse or not.
>> - if you use barrier - both in the host and the guest.
>
> I noticed that when btrfs is mounted with default options, when writing
> i.e. 10 GB on the KVM guest using qcow2 image, 20 GB are written on the
> host (as measured with "iostat -m -p").
>
>
> With ext4 (or btrfs mounted with nodatacow), 10 GB write on a guest
> produces 10 GB write on the host.
>
Whoa 20gb? That doesn't sound right, COW should just mean we get quite a bit of
fragmentation, not write everything twice. What exactly is qemu doing? Thanks,
Josef
next prev parent reply other threads:[~2010-08-30 0:11 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-29 19:34 Tomasz Chmielewski
2010-08-30 0:14 ` Josef Bacik [this message]
2010-08-30 15:59 ` K. Richard Pixley
2010-08-31 21:46 ` Mike Fedyk
2010-08-31 22:01 ` K. Richard Pixley
[not found] ` <4C7D7B14.9020008@noir.com>
2010-09-02 0:18 ` Ted Ts'o
2010-09-02 16:36 ` K. Richard Pixley
2010-09-02 16:49 ` K. Richard Pixley
-- strict thread matches above, loose matches on Subject: below --
2010-07-12 5:24 Giangiacomo Mariotti
2010-07-12 5:54 ` Justin P. Mattock
2010-07-12 7:09 ` Michael Tokarev
2010-07-12 7:17 ` Justin P. Mattock
2010-07-12 13:15 ` Giangiacomo Mariotti
2010-07-12 13:34 ` Giangiacomo Mariotti
2010-07-12 13:40 ` Michael Tokarev
2010-07-12 13:43 ` Josef Bacik
2010-07-12 13:42 ` Michael Tokarev
2010-07-12 13:49 ` Josef Bacik
2010-07-12 20:23 ` Giangiacomo Mariotti
2010-07-12 20:24 ` Josef Bacik
2010-07-13 4:29 ` Avi Kivity
2010-07-14 2:39 ` Giangiacomo Mariotti
2010-07-14 19:49 ` Christoph Hellwig
2010-07-17 5:29 ` Giangiacomo Mariotti
2010-07-17 10:28 ` Ted Ts'o
2010-07-18 7:15 ` Christoph Hellwig
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=20100830001441.GA838@dhcp231-156.rdu.redhat.com \
--to=josef@redhat.com \
--cc=gg.mariotti@gmail.com \
--cc=hch@infradead.org \
--cc=justinmattock@gmail.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mangoo@wpkg.org \
--cc=mjt@tls.msk.ru \
--cc=tytso@mit.edu \
/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
all inboxes | Powered by JetHome®