eugene@eos.UUCP (Eugene Miya) (08/13/88)
No this is not an article about the Sequent Balance 8000,
but I noted the note about defining balance. There's this neat
figure for the high-speed computing conference: Algorithms,
Architectures, Software. I've been thinking about triads like this.
What it boils down to is how were are approaching the "Cray on a
Desk." It means 3 things:
1) CPUs reaching Cray clock rates (near riscy stuff, right,
but most acknowledge soon.
2) Bigger memories. The Original Cray-1, George informed me
had 1 MW memory or about 8 MB which you can now get in
a Mac II [I believe], otherwise there were smaller Crays
with .5 MW. So in memory capacity, we have a Cray on the
desk now.
3) I/O. The biggest bottleneck. I know few who can achieve
Cray Channel speeds: not DEC, not IBM, this is why micros
end up being slower than mainframes, still, but I don't
see much relief in this area yet. We are too CPU infatuated.
Anyways, time to move back further into the undefineable ether.
In 1984 Dan Ingalls proposed the Cray on a wrist [quick look down to
the digital calculator on your wrist ;-)]. [ACM National Mtg, Sept. 84].
Another gross generalization from
--eugene miya, NASA Ames Research Center, eugene@aurora.arc.nasa.gov
resident cynic at the Rock of Ages Home for Retired Hackers:
"Mailers?! HA!", "If my mail does not reach you, please accept my apology."
{uunet,hplabs,ncar,decwrl,allegra,tektronix}!ames!aurora!eugene
"Send mail, avoid follow-ups. If enough, I'll summarize."