[Beowulf] A start in Parallel Programming?

[Peter St. John]

Well I like the vapor-deposition diamond thing. You'd slice the diamond into
thin wafers and treat it just like silicon, except it conducts heat way
better. I think it's already cheap enough to trivialize the cost of the
thickness you'd want for a chip, but they are working on making usefully
wide chunks so you can have big wafers for mass manufacturing just like
silicon.

I think we're zooming to the future, but there is a wall in our future.
Heisenberg will limit how many FLOPS we can do per chunk of stuff, even if's
it Legendary Dark Matter in the Vacuum, and Einstein will limit
communicating results from Proxima Centauri even after it's been converted
to pure Heisenberg Max Dense Supercomputer. The funny thing is that we may
experience this collision in our lifetimes, because we're going so fast and
the rate we are quicking, etc; so I'm an optimistic (like Kurzweil) on the
local scale, but a pessimist on the global one. S curves may all break, but
they converge to a limiting S curve.

History is full of disappointing limitations (you can't express the square
root of two as a quotient of intgers?? But all numbers are quotients of
integers!!) only to discover we can live with it (ok, so we do arithmetic
with irrationals, d'uh). I think converting the mass of Jupitor to
(approximately) maximum deliverable FLOPS per gram will be enough to keep me
amused modelling. Of course the competition for CPU time will
be....astronomical.

Peter


On 3/14/07, Joe Landman <landman at scalableinformatics.com> wrote:
>
> Robert G. Brown wrote:
>
> > So I'm not holding my breath on ML running out this week or next.  I'm
> > more interested in speculating on when the next massive superML jump on
> > TOP of ML will occur, when the next phase/paradigm shift is due that
>
> Ah....
>
> Ok, my concern is that silicon may be running out of room.  Now if we
> all just put on our happy faces, and use GaAs, we can go much faster.
> Yeah, it has some issues ...  But think of how much faster we can be.
> (that and my thesis work might suddenly have applicability, but that is
> another matter).
>
> More seriously, our computing industry can build silicon really well.
> As it keeps shrinking problems arise.  Some of these problems can be
> ameliorated by doing a material paradigm shift.  But our computing
> industry does silicon really well ...
>
> Joe
>
>
> --
>
> Joseph Landman, Ph.D
> Founder and CEO
> Scalable Informatics LLC,
> email: landman at scalableinformatics.com
> web  : http://www.scalableinformatics.com
> phone: +1 734 786 8423
> fax  : +1 734 786 8452 or +1 866 888 3112
> cell : +1 734 612 4615
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.scyld.com/pipermail/beowulf/attachments/20070314/50aa32e2/attachment.html