DOE’s Koonin Focuses on Leadership as Chinese Accelerate Efforts

In late October the New York Times reported that China has built a machine capable of 2.5 petaFLOPS. The Tianhe-1A supercomputer is now the world’s most powerful, 40% faster than the fastest American machine located at Oak Ridge National Laboratory.

Of course this wasn’t a surprise, and the DOE has been publicly positioning itself for this eventuality since at least June when Steven Koonin, the Undersecretary for Science in the DOE, articulated the need for the US to maintain its world leadership position in supercomputing. And since China’s announcement, Koonin and the DOE have been busy trying to position themselves as the real leaders-to-watch in large scale computing.

In a blog post the day after the New York Times story, Koonin lends his weight to DOE efforts to be the first to a useful exaFLOPS computer, taking full ownership of the push:

I am spearheading planning within the Department to develop the next generation of supercomputers over the next decade, which will be capable of exaflop-class performance (a factor of 1000 more powerful than today’s most powerful computers). These machines will require paradigm shifts in both hardware and software in order to use significantly less energy per unit of computation and be more resilient against hardware failures.

Koonin also describes a recent Summit held by the DOE during which Secretary Chu re-articulated the department’s desire to facilitate the Holy Grail of supercomputing: mass adoption.

My office recently held a Simulations Summit, where I hosted more than 70 leaders of academia, industry, government, and national research laboratories to discuss policies and plans for bringing science and simulations to bear in support of national competitiveness. During his keynote address at the Summit, Secretary Chu said that “the DOE strategy should be to make simulation part of everyone’s toolbox.”

(You can find presentations from some of the speakers at the summit website.)

So what does it all mean? It was good to see Koonin avoid the language of crisis that he trotted out in his June editorial. This post appears to indicate that the DOE understands that they need to pay attention to the big checks China is writing and, more importantly, to the native chip and interconnect capabilities that its researchers are producing. It also appears that the DOE is pursuing a reasonable course to its exascale goal. But there are a lot of hurdles between here and there, and it is by no means clear who (if anyone) will reach the finish line first.