Supercomputing: An Industry in Need of a Revolution

M

Mentifex

Bartosz Milewski in his Programming Cafe weblog
proclaims that supercomputing nowadays is
"tedious, uninspiring, and boring" in the same way
that working at Microsoft is now so profit-driven
and boring that "there is an underground pipeline
from Microsoft to Google through which the talent
keeps leaking out."

Milewski says that "not everything is bleak in
the land of supercomputers" and he mentions
the Chapel Parallel Programming Language at
http://chapel.cray.com and the STE||AR Group at
http://stellar.cct.lsu.edu for Systems for "Technologies,
Emergent Parallelism, and Algorithms Research."

http://bartoszmilewski.wordpress.com/2011/11/21/supercomputing-an-industry-in-need-of-a-revolution
is the blogpost by Milewski which has already
gone viral via Reddit and is prompting responses
like my idea that Artificial Intelligence for the
http://code.google.com/p/mindforth/wiki/SuperComputer
is the Grand Challenge that will bring back the
quondam excitement from the days of Seymour Cray
and the late, great early days of Microsoft.

Mentifex (Arthur)
 
R

Robert Myers

Bartosz Milewski in his Programming Cafe weblog
proclaims that supercomputing nowadays is
"tedious, uninspiring, and boring" in the same way
that working at Microsoft is now so profit-driven
and boring that "there is an underground pipeline
from Microsoft to Google through which the talent
keeps leaking out."

Milewski says that "not everything is bleak in
the land of supercomputers" and he mentions
the Chapel Parallel Programming Language athttp://chapel.cray.comand  the STE||AR Group athttp://stellar.cct.lsu.edufor Systems for "Technologies,
Emergent Parallelism, and Algorithms Research."

http://bartoszmilewski.wordpress.com/2011/11/21/supercomputing-an-ind...
is the blogpost by Milewski which has already
gone viral via Reddit and is prompting responses
like my idea that Artificial Intelligence for thehttp://code.google.com/p/mindforth/wiki/SuperComputer
is the Grand Challenge that will bring back the
quondam excitement from the days of Seymour Cray
and the late, great early days of Microsoft.

I propose that we ban the DoE or its contractors from releasing color
plots for publication. Then it will become transparent that they're
not doing anything useful. That should free up money to build a real
supercomputer.

Robert.
 

Members online

Forum statistics

Threads
473,744
Messages
2,569,482
Members
44,901
Latest member
Noble71S45

Latest Threads

Top