xfs
[Top] [All Lists]

Re: Q: Filesystem block sizes available?

To: Roy Sigurd Karlsbakk <roy@xxxxxxxxxxxxx>
Subject: Re: Q: Filesystem block sizes available?
From: Steve Lord <lord@xxxxxxx>
Date: 13 Nov 2001 08:49:30 -0600
Cc: XFS Mailing list <linux-xfs@xxxxxxxxxxx>
In-reply-to: <Pine.LNX.4.30.0111131537380.1009-100000@xxxxxxxxxxxxxxxxx>
References: <Pine.LNX.4.30.0111131537380.1009-100000@xxxxxxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
On Tue, 2001-11-13 at 08:41, Roy Sigurd Karlsbakk wrote:
> Hi
> 
> What block sizes are available on XFS filesystems? I've seen the note
> 'Filesystem block size' on the features list, and it says it's limited to
> the page size on the given architecture (eg .4k on ia64).
> 
> I'm running a lab at Compaq in Oslo now, and some guy at storage told me
> 4k's a waste, and that the SCSI drives wanted 256kB data blocks to really
> speed it all up. Could XFS help me out? I need REALLY HIGH-SPEED access to
> the drives.... Software or hardware RAID - I don't care... I just need the
> speed.
> 
> Thanks a lot.

Right now the filesystem block size on linux is restricted to the
machine page size. Fixing this has been on the todo list for a
long time now, but other projects keep getting given higher priority.
The first thing to appear would be support for smaller block sizes,
not bigger ones - the problem in linux is that the kernel was really
not designed to provide chunks of memory larger than 1 page with any
reliability (i.e. there is a good chance it will fail).

Anyway, having said that, xfs will do larger I/O than this, files should
get layed out contiguous on disk, and both the read and write path will
end up issuing larger requests to the scsi devices.

Steve

> 
> roy
> 
> --
> Roy Sigurd Karlsbakk, MCSE, MCNE, CLS, LCA
> 
> Computers are like air conditioners.
> They stop working when you open Windows.
-- 

Steve Lord                                      voice: +1-651-683-3511
Principal Engineer, Filesystem Software         email: lord@xxxxxxx


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