qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Chris Friesen <chris.friesen@windriver.com>
To: "Benoît Canet" <benoit.canet@irqsave.net>
Cc: Paolo Bonzini <pbonzini@redhat.com>, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] is there a limit on the number of in-flight I/O operations?
Date: Mon, 25 Aug 2014 09:12:54 -0600	[thread overview]
Message-ID: <53FB5276.7050003@windriver.com> (raw)
In-Reply-To: <20140823075658.GA6687@irqsave.net>

On 08/23/2014 01:56 AM, Benoît Canet wrote:
> The Friday 22 Aug 2014 à 18:59:38 (-0600), Chris Friesen wrote :
>> On 07/21/2014 10:10 AM, Benoît Canet wrote:
>>> The Monday 21 Jul 2014 à 09:35:29 (-0600), Chris Friesen wrote :
>>>> On 07/21/2014 09:15 AM, Benoît Canet wrote:
>>>>> The Monday 21 Jul 2014 à 08:59:45 (-0600), Chris Friesen wrote :
>>>>>> On 07/19/2014 02:45 AM, Benoît Canet wrote:
>>>>>>
>>>>>>> I think in the throttling case the number of in flight operation is limited by
>>>>>>> the emulated hardware queue. Else request would pile up and throttling would be
>>>>>>> inefective.
>>>>>>>
>>>>>>> So this number should be around: #define VIRTIO_PCI_QUEUE_MAX 64 or something like than that.
>>>>>>
>>>>>> Okay, that makes sense.  Do you know how much data can be written as part of
>>>>>> a single operation?  We're using 2MB hugepages for the guest memory, and we
>>>>>> saw the qemu RSS numbers jump from 25-30MB during normal operation up to
>>>>>> 120-180MB when running dbench.  I'd like to know what the worst-case would
>>>
>>> Sorry I didn't understood this part at first read.
>>>
>>> In the linux guest can you monitor:
>>> benoit@Laure:~$ cat /sys/class/block/xyz/inflight ?
>>>
>>> This would give us a faily precise number of the requests actually in flight between the guest and qemu.
>>
>>
>> After a bit of a break I'm looking at this again.
>>
>
> Strange.
>
> I would use dd with the flag oflag=nocache to make sure the write request
> does not do in the guest cache though.

I set up another test, checking the inflight value every second.

Running just "dd if=/dev/zero of=testfile2 bs=1M count=700 
oflag=nocache&" gave a bit over 100 inflight requests.

If I simultaneously run "dd if=testfile of=/dev/null bs=1M count=700 
oflag=nocache&" then then number of inflight write requests peaks at 176.

I should point out that the above numbers are with qemu 1.7.0, with a 
ceph storage backend.  qemu is started with

-drive file=rbd:cinder-volumes/.........

Chris

  reply	other threads:[~2014-08-25 15:13 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-18 14:58 [Qemu-devel] is there a limit on the number of in-flight I/O operations? Chris Friesen
2014-07-18 15:24 ` Paolo Bonzini
2014-07-18 16:22   ` Chris Friesen
2014-07-18 20:13     ` Paolo Bonzini
2014-07-18 22:48       ` Chris Friesen
2014-07-19  5:49         ` Paolo Bonzini
2014-07-19  6:27           ` Chris Friesen
2014-07-19  7:23             ` Paolo Bonzini
2014-07-19  8:45               ` Benoît Canet
2014-07-21 14:59                 ` Chris Friesen
2014-07-21 15:15                   ` Benoît Canet
2014-07-21 15:35                     ` Chris Friesen
2014-07-21 15:54                       ` Benoît Canet
2014-07-21 16:10                       ` Benoît Canet
2014-08-23  0:59                         ` Chris Friesen
2014-08-23  7:56                           ` Benoît Canet
2014-08-25 15:12                             ` Chris Friesen [this message]
2014-08-25 17:43                               ` Chris Friesen
2015-08-27 16:37                                 ` Stefan Hajnoczi
2015-08-27 16:33                               ` Stefan Hajnoczi
2014-08-25 21:50                             ` Chris Friesen
2014-08-27  5:43                               ` Chris Friesen
2015-05-14 13:42                                 ` Andrey Korolyov
2015-08-26 17:10                                   ` Andrey Korolyov
2015-08-26 23:31                                     ` Josh Durgin
2015-08-26 23:47                                       ` Andrey Korolyov
2015-08-27  0:56                                         ` Josh Durgin
2015-08-27 16:48                               ` Stefan Hajnoczi
2015-08-27 17:05                                 ` Stefan Hajnoczi
2015-08-27 16:49                               ` Stefan Hajnoczi
2015-08-28  0:31                                 ` Josh Durgin
2015-08-28  8:31                                   ` Andrey Korolyov
2014-07-21 19:47                       ` Benoît Canet
2014-07-21 21:12                         ` Chris Friesen
2014-07-21 22:04                           ` Benoît Canet
2014-07-18 15:54 ` Andrey Korolyov
2014-07-18 16:26   ` Chris Friesen
2014-07-18 16:30     ` Andrey Korolyov
2014-07-18 16:46       ` Chris Friesen
     [not found] <1000957815.25879188.1441820902018.JavaMail.zimbra@redhat.com>
2015-09-09 18:51 ` Jason Dillaman

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=53FB5276.7050003@windriver.com \
    --to=chris.friesen@windriver.com \
    --cc=benoit.canet@irqsave.net \
    --cc=pbonzini@redhat.com \
    --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).