linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Richard Scobie <richard@sauce.co.nz>
To: Michal Soltys <soltys@ziu.info>
Cc: Ben Martin <monkeyiq@users.sourceforge.net>, linux-raid@vger.kernel.org
Subject: Re: Benchmarks: Linux Kernel RAID vs a Hardware RAID setup
Date: Wed, 16 Jul 2008 08:34:47 +1200	[thread overview]
Message-ID: <487D09E7.1010805@sauce.co.nz> (raw)
In-Reply-To: <487CC660.9030004@ziu.info>

Michal Soltys wrote:

> Also looking back at other benchmarks, stripe_cache_size can have pretty 
> tremendous effect on md raid performance. There're other settings that 
> could matter as well (read ahead, queue depths). It would be interesting 
> to see e.g. raid5 256k chunk comparison, done with those altered ( 
> _especially_ md's stripe_cache_size with some high value like 16384 or 
> 32768), and with stripe-width used in ext3 case (if it wasn't used 
> already).

Agreed. On a 16 SATA drive SAS attached md RAID6, large streaming write 
performance goes from around 110MB/s with the default stripe_cache_size, 
to 600MB/s using 16384.

There are major read performance too, setting readahead on the md device 
to 65536.

Regards,

Richard

  reply	other threads:[~2008-07-15 20:34 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-15 14:06 Benchmarks: Linux Kernel RAID vs a Hardware RAID setup Ben Martin
2008-07-15 14:42 ` Justin Piszcz
2008-07-16  4:14   ` Ben Martin
2008-07-15 15:46 ` Michal Soltys
2008-07-15 20:34   ` Richard Scobie [this message]
2008-07-16  2:34     ` Mr. James W. Laferriere
2008-07-16  2:48       ` Richard Scobie
2008-07-16  2:52         ` Mr. James W. Laferriere
2008-07-16  4:17           ` Keld Jørn Simonsen
2008-07-16  5:50       ` Michal Soltys
2008-07-16  3:36   ` Ben Martin
2008-07-16  3:55     ` Richard Scobie
2008-07-16  7:00       ` Dan Williams
2008-07-16 17:08       ` thomas62186218
2008-07-16 19:34         ` Richard Scobie
2008-07-15 16:39 ` Keld Jørn Simonsen
2008-07-15 16:50   ` thomas62186218
2008-07-15 17:39     ` Keld Jørn Simonsen
2008-07-16  0:01       ` Richard Scobie
2008-07-16  0:20         ` Jon Nelson
2008-07-16  4:06           ` Ben Martin
2008-07-16 15:42           ` Ben Martin
2008-07-16  4:23         ` Keld Jørn Simonsen
2008-07-16  5:18           ` Richard Scobie
2008-07-16  8:17             ` Keld Jørn Simonsen
2008-07-15 17:06   ` Jon Nelson
2008-07-16  3:44     ` Ben Martin
2008-07-15 18:40   ` Brad Campbell
2008-07-15 20:12     ` Keld Jørn Simonsen
2008-07-21 16:55       ` Bill Davidsen
2008-07-23  7:45         ` Keld Jørn Simonsen
2008-07-23 10:29           ` Brad Campbell
     [not found]     ` <487CF499.6080105@harddata.com>
2008-07-15 20:15       ` Keld Jørn Simonsen
2008-07-16  3:58   ` Ben Martin
2008-07-16  4:47     ` Keld Jørn Simonsen
2008-07-21 16:58       ` Bill Davidsen
2008-07-15 19:41 ` Keld Jørn Simonsen
2008-07-16  3:25   ` Ben Martin
2008-07-15 20:40 ` Peter Grandi
2008-07-16  3:38 ` Eric Sandeen
2008-07-16 18:54 ` Alan D. Brunelle
2008-07-17  8:26   ` Ben Martin

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=487D09E7.1010805@sauce.co.nz \
    --to=richard@sauce.co.nz \
    --cc=linux-raid@vger.kernel.org \
    --cc=monkeyiq@users.sourceforge.net \
    --cc=soltys@ziu.info \
    /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).