All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rogier Wolff <R.E.Wolff@BitWizard.nl>
To: linux-kernel@vger.kernel.org
Subject: Fwd: Fwd: VFAT performance.
Date: Thu, 5 Jun 2003 14:18:40 +0200	[thread overview]
Message-ID: <20030605141840.C22252@bitwizard.nl> (raw)


Hi,

For the performance freaks: We're copying some data off a VFAT32
partition. We've opened the drive. (Yes I know, you're not supposed to
do that. "Don't do this at home folks!" :-)

When copying /dev/hda, we were able to achieve 11Mbyte per second: Our
100mpbs ethernet throughput. 

When copying large files off /mnt, we see a performance of about 7Mb
per second. We see the head seek to the FAT about twice per second. This
fits in with: 

	4K bytes of FAT contains 1024 fat entries. 
with a 4K clustersize, that would describe about 4Mbytes worth of data. 
So, at 7Mbytes per second we require a new FAT block twice per second. 

I think that we're loosing the 4Mbytes per second of performance due
to the 4 seeks per second that the drive has to perform. 

The way to fix this would be to be able to assign a higher cache
priority (*) to the blocks in the FAT, and to read more than just 4k
per seek to the FAT.

Just something to keep in mind when fiddling with the code again....

			Roger.




(*) i.e. expire them from the buffer cache less easily than normal
blocks.


-- 
** R.E.Wolff@BitWizard.nl ** http://www.BitWizard.nl/ ** +31-15-2600998 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
* The Worlds Ecosystem is a stable system. Stable systems may experience *
* excursions from the stable situation. We are currently in such an      * 
* excursion: The stable situation does not include humans. ***************

             reply	other threads:[~2003-06-05 12:05 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-06-05 12:18 Rogier Wolff [this message]
2003-06-05 17:54 ` Fwd: Fwd: VFAT performance OGAWA Hirofumi

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=20030605141840.C22252@bitwizard.nl \
    --to=r.e.wolff@bitwizard.nl \
    --cc=linux-kernel@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 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.