eugene@pioneer.arpa (Eugene N. Miya) (06/04/88)
This is another experiment on my part of network communication
summaries.
The question is: when do you predict when 1 Tera FLOPS will be
commercially available? (SUSTAINED, not peak compurtation). I know
of 4 proposed architectures using TERA somewhere in their names
(Hi Burton! And I know Monty doesn't get the net). Remember
an architecture in hand is worth an infinite number on the drawing
board.
Background: we had one of our Mrs. Field's discussions and I took down
some dates predicting. Note: I tend to side with Marty Fout's numbers
which I will summarize but not include here. You have 1 week MAX
to respond.
Format: This is an experiment in observation of summaries. Your answer
should be a single year or word. Please put it in the form:
YEAR:
NEVER
or
YEAR:
19XX
or
YEAR:
20XX
Notes without this preliminary form will be thrown out.
You can follow with separate paragraph explanations (which I may or may
not read. Interesting comments I might post. I'll keep the list of
names and years and when a year passes and we have not reaches 1 TF, we
will hunt you down and skin you (just kidding). The point is 1) sparse
unix like answers minimizing comment and 2) realistic answers (the
hunt we will get you). As always send mail and not post. Respected
architects like Bill Wulf get more weight than others and will be
forgiven their sins for at least trying ;-).
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,ihnp4,decwrl,allegra,tektronix}!ames!aurora!eugene
"Send mail, avoid follow-ups. If enough, I'll summarize."