[rescue] Apple Announces Rackmount Server - WITH IDE HARDWARE RAID

[Joshua D Boyd]

On Wed, May 15, 2002 at 02:39:05PM -0400, Patrick Giagnocavo 717-201-3366 wrote:
> On Wed, May 15, 2002 at 12:45:09PM -0700, Rick Hamell wrote:
> > > Of course, do they even meet their own specs?  My IDE drives run nowhere
> > > close to the 33 or 66 megs they advertise.  I'd be shocked if ATA100 drives
> > > run anywhere near 66mb/s either, but each hard drive has to crank 66mb/s
> > > sustained to meet the Apple spec.
> > 
> > 	You all seem to be forgetting that ATA100 drives are 100MB only in
> > BURST, not sustained. Sustained speeds are usually 25-50% of
> > that (Depending on what else is on the bus, how hot the drives are, and
> > what phase the moon is in.)
> 
> I get 15-20MB sustained from my Maxtor 7200.

So, the Xserver can sustain about 80megs then probably, right?
 
> When comparing against SCSI U160, remember that protocol overhead
> means that you don't get 160 either.

So we just need 4 drives that can sustain 25megs per second, and we will have
beaten what Apple can do, while using only one U160 chanel.  73gig drives 
are $300 while 60gig IDE is only about $100, so if we are matching disk
space, then we need 3times as much money.  If we give up matching capacity,
then we can come pretty darn close to matching price and speed.  If we
got rid of the fitting in 1U limitation, we could easily match speed, 
capacity, and price.

So instead of the famous pick two conundrum (fast, cheap, high capacity) we
have pick 3 (fast, cheap, capacity, size).

-- 
Joshua D. Boyd