From: "Mike Black" <mblack@csi-inc.com>
To: Andy Arvai <arvai@scripps.edu>, linux-raid@vger.kernel.org
Subject: Re: filesystems for large arrays
Date: Tue, 11 Feb 2003 07:21:12 -0500 [thread overview]
Message-ID: <028201c2d1c8$111c7fe0$f6de11cc@black> (raw)
In-Reply-To: 200302110925.BAA08326@astra.scripps.edu
Dual Athlon 2000MP
3G Memory
FiberChannel qla2100 interface.
Almost 2TB
Running for about a year now with no problems. Used for data mining development (a fair# of files).
ext3 filesystem on RAID5
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/md6 1949323256 1563695060 366125948 82% /usr6
md6 : active raid5 sdt1[12] sdl1[0] sdk1[11] sdj1[10] sdi1[9] sdh1[8] sdg1[7] sdf1[6] sde1[5] sdd1[4] sda1[3] sdc1[2] sdb1[1]
1950225024 blocks level 5, 128k chunk, algorithm 2 [12/12] [UUUUUUUUUUUU]
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/md6 6190080 401503 5788577 7% /usr6
Done 6/21/02
Linux 2.4.19-pre10 #1 SMP Tue Jun 11 07:04:12 EDT 2002 i686 unknown
tiobench.pl --size 4000
Size is MB, BlkSz is Bytes, Read, Write, and Seeks are MB/secd . -T
File Block Num Seq Read Rand Read Seq Write Rand Write
Dir Size Size Thr Rate (CPU%) Rate (CPU%) Rate (CPU%) Rate (CPU%)
------- ------ ------- --- ----------- ----------- ----------- -----------
. 4000 4096 1 123.2 98.7% 1.090 1.25% 58.23 83.5% 2.755 2.64%
. 4000 4096 2 116.9 94.4% 1.413 1.85% 57.80 102.% 2.759 3.09%
. 4000 4096 4 117.3 95.9% 1.749 2.31% 56.49 104.% 2.760 3.53%
. 4000 4096 8 116.3 96.0% 2.092 2.77% 48.26 91.1% 2.760 3.84%
----- Original Message -----
From: "Andy Arvai" <arvai@scripps.edu>
To: <linux-raid@vger.kernel.org>
Sent: Tuesday, February 11, 2003 4:25 AM
Subject: filesystems for large arrays
> Hi,
>
> Does anyone have comments about the various filesystems typically used
> with large (terabyte) arrays, regarding performance and reliability?
>
> The three most common filesystems seem to be ext3, reiser and xfs. Ext3
> and reiser are part of the standard linux kernel (not sure about xfs),
> implying that they are fairly robust. I've been using reiser and
> haven't had any problems, but I've heard that when the filesystem gets
> full there may be problems and it's also optimized for many small files
> instead of larger files. Ext3 sounds like it is very robust (since it
> is based on ext2), although I've heard the performance is worse than
> reiser. I've heard some good things about xfs, but have never used it.
>
> If anyone has any real-world experiences or benchmarks I would be
> interested.
>
> Andy
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2003-02-11 12:21 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-02-11 9:25 filesystems for large arrays Andy Arvai
2003-02-11 9:47 ` ÷ÉÔÁÌÉÊ
2003-02-11 12:21 ` Mike Black [this message]
2003-02-11 13:13 ` Norman Schmidt
2003-02-11 13:24 ` Mike Black
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='028201c2d1c8$111c7fe0$f6de11cc@black' \
--to=mblack@csi-inc.com \
--cc=arvai@scripps.edu \
--cc=linux-raid@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).