LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Tomasz Chmielewski <mangoo@wpkg.org>
To: Theodore Tso <tytso@mit.edu>,
Tomasz Chmielewski <mangoo@wpkg.org>,
Andi Kleen <andi@firstfloor.org>,
LKML <linux-fsdevel@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: very poor ext3 write performance on big filesystems?
Date: Mon, 18 Feb 2008 17:16:55 +0100 [thread overview]
Message-ID: <47B9AF77.9040702@wpkg.org> (raw)
In-Reply-To: <20080218151632.GD25098@mit.edu>
Theodore Tso schrieb:
>> Are there better choices than ext3 for a filesystem with lots of hardlinks?
>> ext4, once it's ready? xfs?
>
> All filesystems are going to have problems keeping inodes close to
> directories when you have huge numbers of hard links.
>
> I'd really need to know exactly what kind of operations you were
> trying to do that were causing problems before I could say for sure.
> Yes, you said you were removing unneeded files, but how were you doing
> it? With rm -r of old hard-linked directories?
Yes, with rm -r.
> How big are the
> average files involved? Etc.
It's hard to estimate the average size of a file. I'd say there are not
many files bigger than 50 MB.
Basically, it's a filesystem where backups are kept. Backups are made
with BackupPC [1].
Imagine a full rootfs backup of 100 Linux systems.
Instead of compressing and writing "/bin/bash" 100 times for each
separate system, we do it once, and hardlink. Then, keep 40 copies back,
and you have 4000 hardlinks.
For individual or user files, the number of hardlinks will be smaller of
course.
The directories I want to remove have usually a structure of a "normal"
Linux rootfs, nothing special there (other than most of the files will
have multiple hardlinks).
I noticed using write back helps a tiny bit, but as dm and md don't
support write barriers, I'm not very eager to use it.
[1] http://backuppc.sf.net
http://backuppc.sourceforge.net/faq/BackupPC.html#some_design_issues
--
Tomasz Chmielewski
http://wpkg.org
next prev parent reply other threads:[~2008-02-18 16:17 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-02-18 12:57 Tomasz Chmielewski
2008-02-18 14:03 ` Andi Kleen
2008-02-18 14:16 ` Theodore Tso
2008-02-18 15:02 ` Tomasz Chmielewski
2008-02-18 15:16 ` Theodore Tso
2008-02-18 15:57 ` Andi Kleen
2008-02-18 15:35 ` Theodore Tso
2008-02-20 10:57 ` Jan Engelhardt
2008-02-20 17:44 ` David Rees
2008-02-20 18:08 ` Jan Engelhardt
2008-02-18 16:16 ` Tomasz Chmielewski [this message]
2008-02-18 18:45 ` Theodore Tso
2008-02-18 15:18 ` Andi Kleen
2008-02-18 15:03 ` Theodore Tso
2008-02-19 14:54 ` Tomasz Chmielewski
2008-02-19 15:06 ` Chris Mason
2008-02-19 15:21 ` Tomasz Chmielewski
2008-02-19 16:04 ` Chris Mason
2008-02-19 18:29 ` Mark Lord
2008-02-19 18:41 ` Mark Lord
2008-02-19 18:58 ` Paulo Marques
2008-02-19 22:33 ` Mark Lord
2008-02-27 11:20 ` Tomasz Chmielewski
2008-02-27 20:03 ` Andreas Dilger
2008-02-27 20:25 ` Tomasz Chmielewski
2008-03-01 20:04 ` Bill Davidsen
2008-02-19 9:24 ` Vladislav Bolkhovitin
[not found] <9YdLC-75W-51@gated-at.bofh.it>
[not found] ` <9YeRh-Gq-39@gated-at.bofh.it>
[not found] ` <9Yf0W-SX-19@gated-at.bofh.it>
[not found] ` <9YfNi-2da-23@gated-at.bofh.it>
[not found] ` <9YfWL-2pZ-1@gated-at.bofh.it>
[not found] ` <9Yg6H-2DJ-23@gated-at.bofh.it>
2008-02-19 13:14 ` Paul Slootman
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=47B9AF77.9040702@wpkg.org \
--to=mangoo@wpkg.org \
--cc=andi@firstfloor.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tytso@mit.edu \
--subject='Re: very poor ext3 write performance on big filesystems?' \
/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
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).