xfs
[Top] [All Lists]

Re: mkfs.xfs option for RAID

To: "T.Hattori" <tomo@xxxxxxxxx>
Subject: Re: mkfs.xfs option for RAID
From: Danny Cox <danscox@xxxxxxxxxxxxxx>
Date: Fri, 25 May 2001 07:26:27 -0400
Cc: linux-xfs@xxxxxxxxxxx
Organization: Connex Inc
References: <3B0E11EC.30F45CF7@ppi.co.jp>
Reply-to: dcox@xxxxxxxxxx
Sender: owner-linux-xfs@xxxxxxxxxxx
"T.Hattori" wrote:
> 
> Hi.
> For maximam performance and efficient performance, EXT2 file systems can
> be built with "-b" & "-R stride= " options.
> For instance, to match a normal RAID controller's 64KB default stripe
> size, the command  'mke2fs -b 4096 -R stride=16 <device>' is
> appropriate. ( 4KB * 16 = 64KB)
> 
> Can we tune XFS like this using -r option or others?

        Yes!  In fact, mkfs.xfs "notices" if the block device is LVM or RAID,
and sets it's parameters automatically.  FYI, they are:

                sunit - stripe unit size in 512 byte blocks
                swidth - stripe width size in 512 byte blocks

-- 
"Men occasionally stumble over the truth, but most of them pick 
themselves up and hurry off as if nothing had happened." 
   -- Winston Churchill 

Danny

<Prev in Thread] Current Thread [Next in Thread>