[Beowulf] Pony: not yours.

[Eugen Leitl]

On Thu, May 16, 2013 at 05:18:17PM +0200, Andrew Holway wrote:
> There is going to be a paradigm shift or some new kind of disruptive
> technology is going to pop up before 'exascale' happens.

Exascale will happen with conventional technology, it
will just take a bit longer than 2020. We already have
on-die Si photonics, TSV (hybrid) memory stacking and 
the like which should be able to scale.
 
> Quantum or some shizzle. It will do to clusters what the cluster did
> to IBM and Cray.

It's pretty clear we'll be getting self-assembled 
molecular circuitry and spintronics, and reach Avogadro 
scale computing (a ~mole of live bits) but that's probably 
half a century away.