xfs
[Top] [All Lists]

Re: mkfs.xfs

To: Gaspar Bakos <gbakos@xxxxxxxxxxxxxxx>
Subject: Re: mkfs.xfs
From: David Jez <dave.jez@xxxxxxxxx>
Date: Wed, 11 Feb 2004 14:37:33 +0100
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <Pine.SOL.4.58.0402092229300.4489@antu.cfa.harvard.edu>
References: <Pine.SOL.4.58.0402092229300.4489@antu.cfa.harvard.edu>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.4.1i
On Mon, Feb 09, 2004 at 10:34:42PM -0500, Gaspar Bakos wrote:
> Hi,
  Hi,

> I am creating XFS filesystem on a 750Gb partition, which is made up of
> 4x250Gb WD disks in RAID-5 + 3ware card. System is RH9.0 with 2.4.23
> kernel and xfs patch as of 031010 from the oss.sgi site.

> [...]

> hdparm -tT /dev/sda
> /dev/sda:
>  Timing buffer-cache reads:   128 MB in  0.20 seconds =640.00 MB/sec
>  Timing buffered disk reads:  64 MB in  1.55 seconds = 41.29 MB/sec
> 
> It is OK, but I am not shocked by the performance.
  sda is one disk from array or whole array?
  I think this is normal behaviour. Raid 5 isn't raid0 and is quite slow
(and realy slow for writting operations)...

> Are there any recommended tests to test the filesystem before starting to
> populate it with all that 750Gb of data?
  Stress operations with big files and many small files can help.  Try it
under filesystem and direct disk access.

> 
> All the best
> Gaspar
-- 
-------------------------------------------------------
  David "Dave" Jez                Brno, CZ, Europe
 E-mail: dave.jez@xxxxxxxxx
PGP key: finger xjezda00@xxxxxxxxxxxxxxxx
---------=[ ~EOF ]=------------------------------------


<Prev in Thread] Current Thread [Next in Thread>
  • mkfs.xfs, Gaspar Bakos
    • Re: mkfs.xfs, David Jez <=