From: Avi Kivity <avi@redhat.com>
To: Chris Wright <chrisw@sous-sol.org>
Cc: Badari Pulavarty <pbadari@us.ibm.com>,
Christoph Hellwig <hch@infradead.org>,
kvm@vger.kernel.org
Subject: Re: [RFC] vhost-blk implementation
Date: Mon, 29 Mar 2010 23:37:09 +0300 [thread overview]
Message-ID: <4BB10F75.2000701@redhat.com> (raw)
In-Reply-To: <20100329182010.GM1744@sequoia.sous-sol.org>
On 03/29/2010 09:20 PM, Chris Wright wrote:
> * Badari Pulavarty (pbadari@us.ibm.com) wrote:
>
>> I modified my vhost-blk implementation to offload work to
>> work_queues instead of doing synchronously. Infact, I tried
>> to spread the work across all the CPUs. But to my surprise,
>> this did not improve the performance compared to virtio-blk.
>>
>> I see vhost-blk taking more interrupts and context switches
>> compared to virtio-blk. What is virtio-blk doing which I
>> am not able to from vhost-blk ???
>>
> Your io wait time is twice as long and your throughput is about half.
> I think the qmeu block submission does an extra attempt at merging
> requests. Does blktrace tell you anything interesting?
>
It does. I suggest using fio O_DIRECT random access patterns to avoid
such issues.
--
Do not meddle in the internals of kernels, for they are subtle and quick to panic.
next prev parent reply other threads:[~2010-03-29 20:37 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-23 1:00 [RFC] vhost-blk implementation Badari Pulavarty
2010-03-23 1:16 ` Anthony Liguori
2010-03-23 1:45 ` Badari Pulavarty
2010-03-23 2:00 ` Anthony Liguori
2010-03-23 2:50 ` Badari Pulavarty
2010-03-23 10:05 ` Avi Kivity
2010-03-23 14:48 ` Badari Pulavarty
2010-03-23 10:03 ` Avi Kivity
2010-03-23 14:55 ` Badari Pulavarty
2010-03-23 16:53 ` Avi Kivity
2010-03-24 20:05 ` Christoph Hellwig
2010-03-25 6:29 ` Avi Kivity
2010-03-25 15:48 ` Christoph Hellwig
2010-03-25 15:51 ` Avi Kivity
2010-03-25 15:00 ` Asdo
2010-04-05 19:59 ` Christoph Hellwig
2010-04-07 0:36 ` [RFC] vhost-blk implementation (v2) Badari Pulavarty
2010-04-07 0:36 ` Badari Pulavarty
2010-03-23 10:09 ` [RFC] vhost-blk implementation Eran Rom
2010-03-24 20:04 ` Christoph Hellwig
2010-03-24 20:22 ` Badari Pulavarty
2010-03-25 7:57 ` Avi Kivity
2010-03-25 14:36 ` Badari Pulavarty
2010-03-25 15:57 ` Christoph Hellwig
2010-03-26 18:53 ` Eran Rom
2010-04-08 16:17 ` Stefan Hajnoczi
2010-04-05 19:23 ` Christoph Hellwig
2010-04-05 23:17 ` Badari Pulavarty
2010-03-24 20:27 ` Badari Pulavarty
2010-03-29 15:41 ` Badari Pulavarty
2010-03-29 18:20 ` Chris Wright
2010-03-29 20:37 ` Avi Kivity [this message]
2010-03-29 22:51 ` Badari Pulavarty
2010-03-29 23:56 ` Chris Wright
2010-03-30 12:43 ` Avi Kivity
2010-04-05 14:22 ` Stefan Hajnoczi
2010-04-06 2:27 ` Badari Pulavarty
-- strict thread matches above, loose matches on Subject: below --
2010-03-23 0:34 Badari Pulavarty
2010-03-23 12:39 ` Michael S. Tsirkin
2010-03-23 14:47 ` Badari Pulavarty
2010-03-23 17:57 ` Badari Pulavarty
2010-03-23 18:06 ` Michael S. Tsirkin
2010-03-23 19:55 ` Badari Pulavarty
2010-03-24 9:52 ` Michael S. Tsirkin
2010-03-24 11:23 ` Michael S. Tsirkin
2010-03-24 17:58 ` Badari Pulavarty
2010-03-24 18:28 ` Michael S. Tsirkin
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=4BB10F75.2000701@redhat.com \
--to=avi@redhat.com \
--cc=chrisw@sous-sol.org \
--cc=hch@infradead.org \
--cc=kvm@vger.kernel.org \
--cc=pbadari@us.ibm.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.