NVIDIA Lands $12-Million Contract For Exascale Research

As powerful as today's super computers are, it's widely believed that we'll see a tectonic shift in supercomputing in this decade - a shift that will leave the world's current super computers far behind. That change is exascale computing technology, which will be thousands of times faster than petaflop super computers. Today, NVIDIA announced that it has a $12.4-million contract to further its exascale computing research.

Several U.S. government agencies have an interest in this kind of computing power, for reasons as varied as national defense, medical research, and engine improvement. The Department of Energy has been one of the most visible proponents of making exascale computing a reality and it's the agency that awarded NVIDIA the two-year contract, via its FastForward program.

Kepler SMX technology provides more cores than the venerable Fermi shader multiprocessor (SM).
Also, control logic is reduced, making for more room and less power consumption.

So, if exascale computing is so important to achieve, what's the holdup? Electrical power, for one thing. NVIDIA chief scientist Bill Dally described it this way in a blog post today: "One of the great challenges in developing such systems is in making them energy efficient. Theoretically, an exascale system could be built with x86 processors today, but it would require as much as 2 gigawatts of power — the entire output of the Hoover Dam."

NVIDIA Kepler GPU's Dynamic Parallelism

Dynamic Parallelism lets the GPU create new threads without going back to the CPU,
It makes parallel processing more accessible to developers.

That's where NVIDIA comes in. It has already developed processors designed with exascale computing in mind. In fact, according to Dally, NVIDIA's Kepler K20 processors can already reduce an exascale system's energy consumption to 150 megawatts. Kepler K20 processors implement technologies such as SMX, which manages control logic to improve performance, and Dynamic Parallelism, an efficiency tool that makes use of graphics processors. The money from today's contract will help NVIDIA bring power consumption to under 20 megawatts before 2020.

realneil 2 years ago

Congrats to NVIDIA for this. I wish them success attaining their goals too. Trickle down effects should really change the desktop video card market too. I look forward to using them,........(if I'm still alive to see it)

mhenriday 2 years ago

Here's a brief notice about one of the techniques that hopefully will contribute to rendering exascale computing possible : http://www.technologyreview.com/news/428466/making-wiring-that-doesnt-trip-up-computer-chips/?nlid=nldly&nld=2012-07-11. After all, making better chips is useless, if one can't connect them. Amazingly enough, the technique that's described here - heating the chips so that the deposited copper will flow into all the holes - looks similar to the so-called «»reballing techniques that sometimes are successful in resuscitating a motherboard that's died due to overheating. Nothing new under the sun - which is what makes patenting ideas, as opposed to devices, so absurd....


Post a Comment
or Register to comment