Cluster Challenge: Purdue, ITaP student supercomputing team awaits results
Purdue’s student supercomputing team has completed the data-crunching phase of the 2011 Cluster Challenge at SC11, the world’s largest supercomputing conference, and is waiting to see where it will land in the final standings for the international competition.
The deadline for the team of six undergraduates to run as much simulated scientific data from competition organizers as possible on their the student-built and -run supercomputer passed on Wednesday evening (Nov. 16).
The data collection was too massive for any team to process fully. The students are judged on how much they get done, the precision of their results and their ability to explain to a roving group of judges the strategy they used in the Cluster Challenge. The winner will be announced at a luncheon Thursday (Nov. 17) in Seattle, where SC11 is being held this week.
“This year we had a different approach to running the applications,” says Joad Fattah, a junior in computer science from Carmel. “We gave so much time to one application and then moved on to another application.”
In previous years, the team has run multiple applications, something their supercomputer, a hundred times more powerful than a typical personal computer, is easily capable of doing. But concentrating on one application at a time allows the team to bring the machine’s full power to bear on the task.
The Purdue team was one of eight qualifiers for the 2011 Cluster Challenge, one of just four from the U.S. The team is competing against teams from China, Russia and other countries. Purdue’s is the lone team from the Big Ten.
The other Cluster Challenge team members are Alex Bartol, a senior in computer science from Fort Wayne; John Blaas, a senior in computer and information technology from Lafayette; Michael Heffernan, a senior in computer science from Kokomo; Andrew Huff, a junior in computer science from Cary, N.C.; and Tyler Reid, a junior in computer science from Zionsville.
The team members shared the load monitoring and adjusting their machines around the clock over three days. The 2011 applications are used for studying the actions of chemical molecules, the colliding and merging of galaxies, the basic workings of biological life and the motion of the oceans.
The ocean simulator, Parallel Ocean Program, or POP, proved to be an advantage for the Purdue team, whose faculty advisor, earth and atmospheric sciences Professor Mike Baldwin, uses the software in his research. In addition, all the students are taking a high-performance computing class taught by Baldwin this semester.
“He went into great detail on POP,” Blaas said, “so there’s a lot of knowledge transfer.”
Intel is Purdue’s partner in the Cluster Challenge and with 160 of the company’s processors inside the 2011 entry is akin to a mini version of the Hansen cluster supercomputer Purdue installed over the summer and the four other clusters ITaP has built in partnership with Purdue Faculty since 2008. Most of this year’s Purdue team helped build Hansen as student workers at the Rosen Center for Advanced Computing, ITaP’s research computing unit.
Writer: Greg Kline, ITaP science and technology writer, 765-494-8167 (office), 765-426-8545 (mobile), email@example.com