Fact: According to a WIRED report earlier this week, “A new crop of supercomputers is breaking down the petaflop speed barrier, pushing high-performance computing into a new realm that could change science more profoundly than at any time since Galileo.”
Analysis: I wrote a couple of weeks ago about the upcoming (this week) SC08, the big annual Supercomputing conference down in Austin. The big news today was the announcement (just this afternoon) of the winner of the annual Gordon Bell Award:
A team led by Thomas Schulthess of the U.S. Department of Energy’s Oak Ridge National Laboratory received the prestigious 2008 Association for Computing Machinery (ACM) Gordon Bell Prize Thursday after attaining the fastest performance ever in a scientific supercomputing application. – ORNL official press release
I visited Oak Ridge National Labs several times while in government and have recently met with ORNL’s CIO, Scott Studham, to discuss some teragrid research, and knowing the quality of the entire ORNL team I couldn’t be happier for them all with this prize. The nation’s string of national labs is perhaps our finest, most under-appreciated (and underfunded) research jewels, and their contributions to national security and the nation’s scientific future are incalculable. As a nation, the U.S. remains a superpower as a direct result of the work of the national labs.
The award itself is a high honor; it’s named for the legendary supercomputing pioneer Gordon Bell, now a principal researcher in Microsoft Research – his home page here. Here’s more on why ORNL won, again from their press release:
Schulthess is group leader of ORNL’s Computational Materials Science Group … He and colleagues Thomas Maier, Michael Summers and Gonzalo Alvarez, all of ORNL, achieved 1.352 quadrillion calculations a second–or 1.352 petaflops–on ORNL’s Cray XT Jaguar supercomputer with a simulation of superconductors, or materials that conduct electricity without resistance. By modifying the algorithms and software design of its DCA++ code to maximize speed without sacrificing accuracy, the team was able to boost performance tenfold with the help of John Levesque and Jeff Larkin of Cray Inc.
From WIRED again, more on the context of the importance of Jaguar:
Supercomputing has made huge advances over the last decade or so, gradually packing on the ability to handle more and more data points in increasingly complex ways. It has enabled scientists to test theories, design experiments and predict outcomes as never before. But now, the new class of petaflop-scale machines is poised to bring about major qualitative changes in the way science is done.
“The new capability allows you to do fundamentally new physics and tackle new problems,” said Thomas Zacharia, who heads up computer science at Oak Ridge National Laboratory in Tennessee, home of the second place Cray XT5 Jaguar supercomputer. “And it will accelerate the transition from basic research to applied technology.”
Oh – and a final bit of news from the Supercomputing 2008 conference: there’s a new entrant in the list of the Top 10 world’s most powerful supercomputers: Microsoft! Partnering with Shanghai Supercomputer Center and Dawning Information Industry Co. Ltd., their entry ranked at No. 10 with 180.6 teraflops of parallel computing speed, and 77.5 percent efficiency.
That’s being taken as a surprising achievement considering that twelve months ago Microsoft was at 116 on the Top500 list (maintained at www.Top500.org). It’s testimony that the Windows HPC Server 2008 software can combine with cool hardware (like our alliance with Cray) to deliver a mini-supercomputer at a bargain price.
So even if you can’t afford a petaflop machine, you have an option🙂
Filed under: Government, Microsoft, R&D, Technology Tagged: | ACM, Austin, computer, computers, computing, Cray, Dawning, DOE, Gordon Bell, HPC, Microsoft, National Lab, national laboratories, national labs, Oak Ridge, ORNL, SC08, Scott Studham, Shanghai Supercomputer Center, SSC, supercomputer, supercomputing, tech, Technology, Thomas Schulthess, Top500, Windows HPC Server, Wired