Every six months, the folks at Top500.org put together what has become the most-recognized metric of supercomputing speed and power, the Top 500 list. While there’s ongoing debate about the meaning and value of a top 500 ranking, it’s proven to be the most often-cited guide to where the “big iron” really is — touted by vendors, researchers, agencies, even policymakers as a way to demonstrate their high-performance computing capabilities.
The newest ranking, released at this week’s SC07 conference in Reno, Nevada, struck me as noteworthy because of the sites ranked in the Top 5. In the last ranking back in June 2007, the top 5 sites were all in the U.S. — 4 DOE Labs and IBM’s Watson Research Center. In the latest ranking, only two U.S. sites rank in the top 5 — DOE’s Lawrence Livermore National Lab (it’s BlueGene/L machine is #1) and the New Mexico Computing Applications Center (an SGI machine at #3). The other three are Germany’s Foshungszentrum Juelich (another BlueGene machine at #2), India’s Computational Research Laboratories (an HP cluster at #4), and a classified machine at a Swedish government agency (another HP cluster at #5).
I don’t want to draw any huge conclusions from this about the state of U.S. HPC efforts — after all, all the machines in the top 5 (indeed in the Top 15) are manufactured by U.S.-based companies (though the Indian machine apparently makes use of their own “innovative routing technology”). But if nothing else, this appears to demonstrate a commitment by these other countries — all competitors in the global economy — to really strong investments in HPC technologies. It represents further capacity-building on their part, the recognition that in order to compete, they must compute (to steal a catchphrase from the Council on Competitiveness).
As we look for ways to benchmark U.S. competitiveness and judge where future trends will take us, taking note of our competitors investments in high-performance computing ought to factor in pretty heavily, I think. That said, the U.S. continues to do pretty well in investing in HPC.
Last July, or maybe even a bit earlier, the members of the Task Force on the Future of American Innovation — an organization of high-tech companies and academic societies (CRA is a member) devoted to increasing federal support for basic research in the physical sciences, mathematics, computing and engineering — looked ahead at the calendar and realized that in a few short months the anniversary of one of the most significant events in world history would be upon them. The launch of Sputnik on October 4, 1957 changed the landscape for science and engineering in the U.S. by forcing a focus on improvements to our science and engineering to “close the gap” with the Soviets (whether there really was a gap at the time…a subject for someone else’s post). Sputnik led to a significant investment in space-related research and engineering, of course, but it also — and maybe more importantly — led to a overall buttressing of the science and engineering ecosystem in the United States. The payoff of that support over the last 50 years has been the United States’ leadership position in the global economy, the high standard of living for our citizens, and the dominance of our military. Given the focus of the Task Force, it seemed appropriate to find ways to commemorate the anniversary and the launch’s impact.
One of the ideas tossed around was that the Task Force should get a little Web 2.0-ish and sponsor a video contest on YouTube: create a 3-minute video showcasing how federally funded research has changed American life. The winner would get $1,000, plus an all-expense paid trip to Washington, DC, to watch as their video was played a Capitol Hill event.
Now, I like to think that I’m a pretty web-savvy guy — I twitter; I’ve got a TumbleLog — but I wasn’t at all convinced that this contest would amount to much at all. I could only imagine the sort of the entries that this contest, once unleashed on the YouTube community, might inspire. I was less convinced after my colleagues on the Task Force created the video announcing the contest. I respect the Task Force members as ardent and effective advocates for science, but, uh, videographers we’re not. And the first entry to the contest didn’t give me much hope.
But we let the contest play out and, remarkably, some good stuff started coming in. A few teams really took some time to come up with interesting approaches. And the eventual contest winner’s was just outstanding:
(The director/creator of the video is Adan Vielma, a Junior at Lewis and Clark College in Oregon. He says he spent only 14 hours creating the images and putting it all together.)
Vielma attended a screening of his video at a November 8th briefing of the Congressional R&D Caucus, hosted by Representatives Rush Holt (D-NJ) and Judy Biggert (R-IL), as part of an event called “Sputnik in a YouTube Age.” The event, sponsored by the Task Force, featured remarks by two former NASA astronauts — Mae Jameison, the first African-American woman in space, and Kathryn Sullivan, the first U.S. woman to spacewalk — focusing on how Sputnik marked the beginning of an investment in science and math that led to the greatest explosion of scientific advancement the world had ever seen. The event, held on an otherwise busy Thursday, was absolutely packed, and I have to think a large part of the draw was the novelty of seeing the YouTube submissions.
So maybe we’ll have to explore other Web 2.0-ish ways of making our case…. A Googlemap mashup of innovation? Best innovation-oriented Facebook App? We’ll see.
“Veteran supercomputing researcher” and current CRA Board Chair Dan Reed, will leave his position at the University of North Carolina’s Renaissance Computing Institute (RENCI) and join Microsoft Research on December 3rd, he announced today. From his blog post:
On December 3, I will embark on the next installment of my own future, which will place me in the center of the ever-evolving computing revolution. On that day, I will be joining Microsoft to head a new research initiative (see the Microsoft Research press release and RENCI/UNC press release) in scalable and multicore computing. I am enormously excited, as these are among the most interesting technical problems in computing, and they are my long-time professional interests. I will be working with Microsoft researchers and product developers, as well as industry partners and academics. It doesnt get any cooler than this.
Check the post for a bunch more detail on the move. Congrats, Dan!
From the CCC Blog
- DARPA Issues PERFECT Solicitation Erwin Gianchandani
- Microsoft Announces New Policy-Focused Internship Program Erwin Gianchandani
- NAE Elects its Class of 2012 Erwin Gianchandani
- NIST to Fund Pilot Projects Advancing Trusted Identities Erwin Gianchandani