From: Christoph Lameter <clameter@sgi.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, hch@infradead.org
Subject: Re: [patch 00/14] Page cache cleanup in anticipation of Large Blocksize support
Date: Thu, 14 Jun 2007 19:04:27 -0700 (PDT) [thread overview]
Message-ID: <Pine.LNX.4.64.0706141855570.3571@schroedinger.engr.sgi.com> (raw)
In-Reply-To: <20070614184013.1ff51d34.akpm@linux-foundation.org>
On Thu, 14 Jun 2007, Andrew Morton wrote:
> There will be files which should use 64k but which instead end up using 4k.
>
> There will be files which should use 4k but which instead end up using 64k.
>
> Because determining which size to use requires either operator intervention
> or kernel heuristics, both of which will be highly unreliable.
>
> It's better to just make 4k pages go faster.
Initially its quite easy to have a filesystem for your 4k files (basically
the distro you are running) and an archive for video / audio etc files
that has 64k size for data. In the future filesystem may support sizes set
per directory. Basically if things get to slow you can pull the lever.
> > Magical? There is nothing magical about doing transfers in the size that
> > is supported by a device. That is good sense.
>
> By magical heuristics I'm referring to the (required) tricks and guesses
> which the kernel will need to deploy to be able to guess which page-size it
> should use for each file.
>
> Because without such heuristics, none of this new stuff which you're
> proposing would ever get used by 90% of apps on 90% of machines.
In the patchset V3 one f.e. simply formats a volume by specifying the
desired blocksize. If one gets into trouble with fsck and other slowdown
associated with large file I/O then they are going to be quite fast to
format a partition with larger blocksize. Its a know technology in many
Unixes.
The approach essentially gives one freedom to choose a page size. This is
a tradeoff between desired speed, expected file sizes, filesystem behavior
and acceptable fragmentation overhead. If we do this approach then I think
we will see the mkfs.XXX tools to automatically make intelligent choices
on which page size to use. They are all stuck at 4k at the moment.
> > Of course there is. The seeks are reduced since there are an factor
> > of 16 less metadata blocks. fsck does not read files. It just reads
> > metadata structures. And the larger contiguous areas the faster.
>
> Some metadata is contiguous: inode tables, some directories (if they got
> lucky), bitmap tables. But fsck surely reads them in a single swoop
> anyway, so there's no gain there.
The metadata needs to refer to 1/16th of the earlier pages that need to be
tracked. metadata is shrunk significantly.
> Other metadata (indirect blocks) are 100% discontiguous, and reading those
> with a 64k IO into 64k of memory is completely dumb.
The effect of a larger page size is that the filesystem will
place more meta data into a single page instead of spreading it out.
Reading a mass of meta data with a 64k read is an intelligent choice to
make in particular if there is a large series of such reads.
next prev parent reply other threads:[~2007-06-15 2:04 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-06-14 19:38 [patch 00/14] Page cache cleanup in anticipation of Large Blocksize support clameter
2007-06-14 19:38 ` [patch 01/14] Define functions for page cache handling clameter
2007-06-14 19:56 ` Sam Ravnborg
2007-06-14 19:58 ` Christoph Lameter
2007-06-14 20:07 ` Sam Ravnborg
2007-06-14 19:38 ` [patch 02/14] Pagecache zeroing: zero_user_segment, zero_user_segments and zero_user clameter
2007-06-14 19:38 ` [patch 03/14] Use page_cache_xx function in mm/filemap.c clameter
2007-06-14 19:38 ` [patch 04/14] Use page_cache_xxx in mm/page-writeback.c clameter
2007-06-14 19:38 ` [patch 05/14] Use page_cache_xxx in mm/truncate.c clameter
2007-06-14 19:38 ` [patch 06/14] Use page_cache_xxx in mm/rmap.c clameter
2007-06-14 19:38 ` [patch 07/14] Use page_cache_xx in mm/filemap_xip.c clameter
2007-06-14 19:38 ` [patch 08/14] Use page_cache_xx in mm/migrate.c clameter
2007-06-14 19:38 ` [patch 09/14] Use page_cache_xx in fs/libfs.c clameter
2007-06-14 19:38 ` [patch 10/14] Use page_cache_xx in fs/sync clameter
2007-06-14 19:38 ` [patch 11/14] Use page_cache_xx in fs/buffer.c clameter
2007-06-14 19:38 ` [patch 12/14] Use page_cache_xxx in mm/mpage.c clameter
2007-06-14 19:38 ` [patch 13/14] Use page_cache_xxx in mm/fadvise.c clameter
2007-06-14 19:38 ` [patch 14/14] Use page_cache_xx in fs/splice.c clameter
2007-06-14 20:06 ` [patch 00/14] Page cache cleanup in anticipation of Large Blocksize support Andrew Morton
2007-06-14 21:07 ` Christoph Hellwig
2007-06-14 21:25 ` Dave McCracken
2007-06-14 21:20 ` Christoph Lameter
2007-06-14 21:32 ` Andrew Morton
2007-06-14 21:37 ` Christoph Lameter
2007-06-14 22:04 ` Andrew Morton
2007-06-14 22:22 ` Christoph Lameter
2007-06-14 22:49 ` Andrew Morton
2007-06-15 0:45 ` Christoph Lameter
2007-06-15 1:40 ` Andrew Morton
2007-06-15 2:04 ` Christoph Lameter [this message]
2007-06-15 2:23 ` Andrew Morton
2007-06-15 2:37 ` Christoph Lameter
2007-06-15 9:03 ` David Chinner
2007-06-14 23:30 ` David Chinner
2007-06-14 23:41 ` Andrew Morton
2007-06-15 0:29 ` David Chinner
2007-06-15 15:05 ` Dave Kleikamp
2007-06-17 1:25 ` Arjan van de Ven
2007-06-17 5:02 ` Matt Mackall
2007-06-18 2:08 ` Christoph Lameter
2007-06-18 3:00 ` Arjan van de Ven
2007-06-18 4:50 ` William Lee Irwin III
2007-06-14 23:54 ` David Chinner
2007-07-02 18:16 ` Badari Pulavarty
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=Pine.LNX.4.64.0706141855570.3571@schroedinger.engr.sgi.com \
--to=clameter@sgi.com \
--cc=akpm@linux-foundation.org \
--cc=hch@infradead.org \
--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 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).