From: Mark Fasheh <mfasheh@suse.com>
To: ocfs2-devel@oss.oracle.com
Subject: [Ocfs2-devel] [PATCH 3/5] ocfs2: use allocation reservations for directory data
Date: Fri, 19 Mar 2010 17:14:25 -0700 [thread overview]
Message-ID: <20100320001425.GL11402@wotan.suse.de> (raw)
In-Reply-To: <20100319224310.GF15539@mail.oracle.com>
On Fri, Mar 19, 2010 at 03:43:10PM -0700, Joel Becker wrote:
> On Tue, Mar 16, 2010 at 11:59:12PM -0700, Mark Fasheh wrote:
> > Use the reservations system for unindexed dir tree allocations. We
> > don't bother with the indexed tree as reads from it are mostly random
> > anyway. By default this behavior is turned off and can be turned on
> > via mount option 'dir_resv'. Workloads which create many large
> > directories will want to turn it on.
> >
> > A future improvement will be to use a different window size for
> > directory inodes. Once done, we should change the default for this
> > feature to be enabled.
>
> Why can't we turn this on and avoid a mount option? Sure, the
> reservation holds out some space, but it will be cannibalized if
> the directory doesn't grow, right?
Yeah, it'll work. My concern was that we'd be cannibalizing those too much
since the window sizes are optimized for file data. It's only a hunch though
that directories might want smaller windows - I didn't get much data on
directory growth since my focus was on file data.
Maybe I should just go ahead and hack it up to know if a reservation is for
a directory and it could provide some fewer number of bits (maybe half the
default)? Adjusting the window size values is something we can do very
easily until we get it right.
Adding a flag field on the resv struct could accomplish this pretty easily,
and I could re-use the flag for the 'temporary' reservations, thus losing a
function call parameter in the process.
I agree though that losing the extra mount option would be good.
--Mark
--
Mark Fasheh
next prev parent reply other threads:[~2010-03-20 0:14 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-17 6:59 [Ocfs2-devel] [PATCH 0/5] Ocfs2 allocation reservations Mark Fasheh
2010-03-17 6:59 ` [Ocfs2-devel] [PATCH 1/5] ocfs2: " Mark Fasheh
2010-03-19 22:40 ` Joel Becker
2010-03-19 23:56 ` Mark Fasheh
2010-03-17 6:59 ` [Ocfs2-devel] [PATCH 2/5] ocfs2: use allocation reservations during file write Mark Fasheh
2010-03-17 6:59 ` [Ocfs2-devel] [PATCH 3/5] ocfs2: use allocation reservations for directory data Mark Fasheh
2010-03-19 22:43 ` Joel Becker
2010-03-20 0:14 ` Mark Fasheh [this message]
2010-03-20 1:25 ` Joel Becker
2010-03-20 3:47 ` Mark Fasheh
2010-03-20 6:18 ` Joel Becker
2010-03-21 23:26 ` Mark Fasheh
2010-03-22 20:41 ` Joel Becker
2010-03-17 6:59 ` [Ocfs2-devel] [PATCH 4/5] ocfs2: allocate btree internal block groups from the global bitmap Mark Fasheh
2010-03-17 6:59 ` [Ocfs2-devel] [PATCH 5/5] ocfs2: remove ocfs2_local_alloc_in_range() Mark Fasheh
2010-03-17 20:17 ` [Ocfs2-devel] [PATCH 0/5] Ocfs2 allocation reservations Mark Fasheh
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=20100320001425.GL11402@wotan.suse.de \
--to=mfasheh@suse.com \
--cc=ocfs2-devel@oss.oracle.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.