All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mark Nelson <mark.nelson@inktank.com>
To: James Harper <james.harper@bendigoit.com.au>
Cc: "ceph-devel@vger.kernel.org" <ceph-devel@vger.kernel.org>
Subject: Re: poor write performance
Date: Fri, 19 Apr 2013 09:50:43 -0500	[thread overview]
Message-ID: <517159C3.5030100@inktank.com> (raw)
In-Reply-To: <6035A0D088A63A46850C3988ED045A4B4D739E99@BITCOM1.int.sbss.com.au>

On 04/19/2013 06:09 AM, James Harper wrote:
> I just tried a 3.8 series kernel and can now get 25mbytes/second using dd with a 4mb block size, instead of the 700kbytes/second I was getting with the debian 3.2 kernel.

That's.... unexpected.  Was this the kernel on the client, the OSDs, or 
both?

>
> I'm still getting 120kbytes/second with a dd 4kb block size though... is that expected?

that's still quite a bit lower than I'd expect as well.  What were your 
fs mount options on the OSDs?  Can you try some rados bench read/write 
tests on your pool?  Something like:

rados -p <pool> -b 4096 bench 300 write --no-cleanup -t 64
rados -p <pool> -b 4096 bench 300 seq -t 64

with 2 drives and 2x replication I wouldn't expect much without RBD 
cache, but 120kb/s is rather excessively bad. :)

>
> James
>

Mark

  reply	other threads:[~2013-04-19 14:50 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-18 11:46 poor write performance James Harper
2013-04-18 12:15 ` Wolfgang Hennerbichler
2013-04-18 23:11   ` James Harper
2013-04-20 10:52     ` Harald Rößler
2013-04-20 11:12       ` James Harper
2013-04-20 21:04         ` Jeff Mitchell
2013-04-18 13:43 ` Mark Nelson
2013-04-18 16:46   ` Andrey Korolyov
2013-04-18 17:01     ` Mark Nelson
2013-04-18 23:23   ` James Harper
2013-04-19  7:21     ` James Harper
2013-04-19  7:30       ` James Harper
2013-04-19 11:09         ` James Harper
2013-04-19 14:50           ` Mark Nelson [this message]
2013-04-20  0:33             ` James Harper
2013-04-20  1:30               ` James Harper
2013-04-21 13:52                 ` Mark Nelson
2013-04-22  5:32                   ` James Harper
2013-04-22 11:34                     ` Mark Nelson
2013-04-22 11:40                       ` James Harper
2013-04-21 17:56               ` Sylvain Munaut
2013-04-21 23:04                 ` James Harper
2013-04-22  8:34                   ` Sylvain Munaut
2013-04-22 11:34                     ` James Harper
2013-04-22 11:39                       ` Mark Nelson
2013-04-22 11:48                         ` James Harper
2013-04-22 12:01                           ` Mark Nelson
2013-04-22 13:47                             ` Mark Nelson
2013-04-22 15:20                         ` Sage Weil
2013-04-22 15:35                           ` Sylvain Munaut

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=517159C3.5030100@inktank.com \
    --to=mark.nelson@inktank.com \
    --cc=ceph-devel@vger.kernel.org \
    --cc=james.harper@bendigoit.com.au \
    /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.