public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Mike Fedyk <mfedyk@matchmail.com>
To: Linus Torvalds <torvalds@osdl.org>
Cc: Helge Hafting <helgehaf@aitel.hist.no>,
	jw schultz <jw@pegasys.ws>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: raid0 slower than devices it is assembled of?
Date: Tue, 16 Dec 2003 12:58:53 -0800	[thread overview]
Message-ID: <20031216205853.GC1402@matchmail.com> (raw)
In-Reply-To: <Pine.LNX.4.58.0312160825570.1599@home.osdl.org>

On Tue, Dec 16, 2003 at 08:42:52AM -0800, Linus Torvalds wrote:
> My personal guess is that modern RAID0 stripes should be on the order of
> several MEGABYTES in size rather than the few hundred kB that most people
> use (not to mention the people who have 32kB stripes or smaller - they
> just kill their IO access patterns with that, and put the CPU at
> ridiculous strain).

Larger stripes may help in general, but I'd suggest that for raid5 (ie, not
raid0), the stripe size should not be enlarged as much.  On many
filesystems, a bitmap change, or inode table update shouldn't require
reading a large stripe from several drives to complete the pairity
calculations.

Probably finding the largest block of data the drive can return in one
command would be a good size for a raid5 stripe.  That's just speculation
though.

  reply	other threads:[~2003-12-16 20:59 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-12-15 13:34 raid0 slower than devices it is assembled of? Witold Krecicki
2003-12-15 15:44 ` Witold Krecicki
2003-12-16  4:01 ` jw schultz
2003-12-16 14:51   ` Helge Hafting
2003-12-16 16:42     ` Linus Torvalds
2003-12-16 20:58       ` Mike Fedyk [this message]
2003-12-16 21:11         ` Linus Torvalds
2003-12-17 10:53           ` Jörn Engel
2003-12-17 11:39           ` Peter Zaitsev
2003-12-17 16:01             ` Linus Torvalds
2003-12-17 18:37               ` Mike Fedyk
2003-12-17 21:55               ` bill davidsen
2003-12-17 17:02             ` bill davidsen
2003-12-17 20:14               ` Peter Zaitsev
2003-12-17 19:22       ` Jamie Lokier
2003-12-17 19:40         ` Linus Torvalds
2003-12-17 22:36           ` bill davidsen
2003-12-18  2:47         ` jw schultz
2003-12-17 22:29       ` bill davidsen
2003-12-18  2:18         ` jw schultz
2004-01-08  4:54       ` Greg Stark
2003-12-16 20:51     ` Andre Hedrick
2003-12-16 21:04       ` Andre Hedrick
2003-12-16 21:46         ` Witold Krecicki
2003-12-16 20:09   ` Witold Krecicki
2003-12-16 21:11   ` Adam Kropelin
2003-12-16 21:25 ` jw schultz

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=20031216205853.GC1402@matchmail.com \
    --to=mfedyk@matchmail.com \
    --cc=helgehaf@aitel.hist.no \
    --cc=jw@pegasys.ws \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@osdl.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