From: "Bryan J. Smith" <thebs413@yahoo.com>
To: Justin Piszcz <jpiszcz@lucidpixels.com>,
Ralf Gross <Ralf-Lists@ralfgross.de>
Cc: linux-xfs@oss.sgi.com
Subject: Re: Re: mkfs options for a 16x hw raid5 and xfs (mostly large files)
Date: Wed, 26 Sep 2007 10:38:33 -0700 (PDT) [thread overview]
Message-ID: <216020.40502.qm@web32910.mail.mud.yahoo.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0709261258180.24646@p34.internal.lan>
Justin Piszcz <jpiszcz@lucidpixels.com> wrote:
> I wonder where the bottleneck lies.
The microcontroller.
Listen, for the last time, hardware RAID is _not_ for non-blocking
I/O. Hardware RAID is for in-line XOR streaming off-load, so it
doesn't tie up a system interconnect (which isn't an ideal use for
it).
A hardware RAID card is when you have other things going on in your
interconnect that you don't want the parity LOAD-XOR-STOR to take
away from what it could be using for the service.
It will _never_ have the "raw performance" of OS optimized software
RAID. At the same time, OS optimized software RAID's impact on the
system interconnect is one of those "unmeasurable" details _unless_
you actually benchmark your application.
I have repeatedly had issues with elementary UDP/IP NFS performance
when the PIO of software RAID is hogging the system interconnect.
Same deal for large numbers of large database record commits.
--
Bryan J. Smith Professional, Technical Annoyance
b.j.smith@ieee.org http://thebs413.blogspot.com
--------------------------------------------------
Fission Power: An Inconvenient Solution
next prev parent reply other threads:[~2007-09-26 17:38 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-09-23 9:38 mkfs options for a 16x hw raid5 and xfs (mostly large files) Ralf Gross
2007-09-23 12:56 ` Peter Grandi
2007-09-26 14:54 ` Ralf Gross
2007-09-26 16:27 ` [UNSURE] " Justin Piszcz
2007-09-26 16:54 ` Ralf Gross
2007-09-26 16:59 ` Justin Piszcz
2007-09-26 17:38 ` Bryan J. Smith [this message]
2007-09-26 17:41 ` Justin Piszcz
2007-09-26 17:55 ` Bryan J. Smith
2007-09-26 17:13 ` [UNSURE] " Bryan J. Smith
2007-09-26 17:27 ` Justin Piszcz
2007-09-26 17:35 ` Bryan J. Smith
2007-09-26 17:37 ` Justin Piszcz
2007-09-26 17:38 ` Justin Piszcz
2007-09-26 17:49 ` Bryan J. Smith
2007-09-27 15:22 ` Ralf Gross
2007-09-24 17:31 ` Ralf Gross
2007-09-24 18:01 ` Justin Piszcz
2007-09-24 20:39 ` Ralf Gross
2007-09-24 20:43 ` Justin Piszcz
2007-09-24 21:33 ` Ralf Gross
2007-09-24 21:36 ` Justin Piszcz
2007-09-24 21:52 ` Ralf Gross
2007-09-25 12:35 ` Ralf Gross
2007-09-25 12:50 ` Justin Piszcz
2007-09-25 13:44 ` Bryan J Smith
2007-09-25 12:57 ` KELEMEN Peter
2007-09-25 13:49 ` Ralf Gross
2007-09-25 14:08 ` Bryan J Smith
2007-09-25 16:07 ` Ralf Gross
2007-09-25 16:28 ` Bryan J. Smith
2007-09-25 17:25 ` Ralf Gross
2007-09-25 17:41 ` Bryan J. Smith
2007-09-25 19:13 ` Ralf Gross
2007-09-25 20:23 ` Bryan J. Smith
2007-09-25 16:48 ` Justin Piszcz
2007-09-25 18:00 ` Bryan J. Smith
2007-09-25 18:33 ` Ralf Gross
2007-09-25 23:38 ` Justin Piszcz
2007-09-26 8:23 ` Ralf Gross
2007-09-26 8:42 ` Justin Piszcz
2007-09-26 8:49 ` Ralf Gross
2007-09-26 9:52 ` Justin Piszcz
2007-09-26 15:03 ` Bryan J Smith
2007-09-26 15:15 ` Ralf Gross
2007-09-26 17:08 ` Bryan J. Smith
2007-09-26 16:24 ` Justin Piszcz
2007-09-26 17:11 ` Bryan J. Smith
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=216020.40502.qm@web32910.mail.mud.yahoo.com \
--to=thebs413@yahoo.com \
--cc=Ralf-Lists@ralfgross.de \
--cc=b.j.smith@ieee.org \
--cc=jpiszcz@lucidpixels.com \
--cc=linux-xfs@oss.sgi.com \
/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.