linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "A. Liemen" <maillist@liemen.net>
To: linux-raid@vger.kernel.org
Subject: Random Seek on Array as slow as on single disk
Date: Sun, 16 Jul 2006 17:37:47 +0200	[thread overview]
Message-ID: <20060716153736.E36C5200967@smtp.funpic.de> (raw)

Hi,

I'm having random seek performance problems on our Raid 5 and Raid 10 
arrays. The random seek performance of the whole 14 disk array is just 
as slow as a single disk. This has been tested with 4 identical boxes, 
Single Disk, Raid1, Raid5, Raid10 configurations. Raid5 / Raid10 has 
been tested with 14 disks arrays.

System:

Debian 2.6.15-1-amd64-k8-smp dual opteron 275
Controller: Areca ARC 1160 PCI-X 1GB Cache
Disks: 14x WD400YR - 400GB SATA
Raid: Raid5 / Raid 10 - 64k / 128k stripe
Filesystems: XFS, Reiser, ext3
Ram: 16gb
Schedulers: deadline, anti, cfq


Benchmark:

./bonnie++ -uroot -s40000M

1-4 threads, overall random seeks stay the same (~360 seeks / sec)

Sample bonnie Raid10 output can be found under:

http://alexander.liemen.net/bonnie_sample.txt


No matter if I benchmark a single disk, Raid1 2 disks, Raid5/10 14 disks 
, single thread / multiple threads,
random seek performance always stays at ~ 340-360 seeks per second.

I have also observed overall bad random seek performance on an identical 
production box so it's not just bonnie producing strange results.


Can anyone help me on this?


Thank you a lot
Alex

             reply	other threads:[~2006-07-16 15:37 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-07-16 15:37 A. Liemen [this message]
2006-07-16 16:39 ` Random Seek on Array as slow as on single disk Jeff Breidenbach
     [not found] ` <20060716163931.D95FE200970@smtp.funpic.de>
2006-07-16 16:49   ` A. Liemen
2006-07-16 18:25     ` Jeff Breidenbach
2006-07-17  7:23     ` Mattias Wadenstein
2006-07-17 17:54     ` Bill Davidsen

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=20060716153736.E36C5200967@smtp.funpic.de \
    --to=maillist@liemen.net \
    --cc=linux-raid@vger.kernel.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).