On Mon, Apr 28, 2003 at 02:42:43PM -0700, Jason Thorpe wrote:
> 
> On Monday, April 28, 2003, at 02:02  PM, Manuel Bouyer wrote:
> 
> >the "queue algorithm modifier" bits of the control mode page, maybe ?
> >SPC-3 says that the default value should be "restricted reordering", 
> >which
> >really means no reordering.
> >I should try changing this, to see if it improve performances.
> >For a large sequential write, I'm not sure it would make much 
> >difference.
> 
> It certainly could make a difference!  If you have a large sequential 
> write, it could allow the disk to do a much larger internal write 
> operation (merging the writes in its own buffer).

In this case I suspect it already gets the requests in increasing
blocks number order.

-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 24 ans d'experience feront toujours la difference
--