On Wed, May 22, 2002 at 09:17:51PM -0700, Wolfgang Rupprecht wrote:
> 
> Manuel Bouyer writes:
> > On Wed, May 22, 2002 at 06:28:18AM -0700, Wolfgang Rupprecht wrote:
> > > Anyone have a good model for what is going on here?
> > IDE doesn't do disconnect/reselect, so when you have 2 concurent one
> > drive is idle while the other one is working. So in your test it's
> > expected to have the per-drive transfer rate drop by half.  I'm not
> > sure why it's more than half. Maybe the extra delay between 2
> > transfers prevents the drive from doing some kind of "streaming" ?
> 
> Thanks!  I didn't know about the lack of disconnect on IDE.  That does
> make one wonder if ATA-100 has any utility.  If drives can't share the
> bus why bother with a channel that can go 2.5x faster than the data
> can come off the disks?

To read data from disk cache ? :)

Newer IDE supports it, but I didn't seriously look at it yet. I wonder how
it can inpact performances (all disconnect/reselect needs host attention,
and will triple the interrupt load by 3).

--
Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
--