New Hansen research supercomputer activated, capacity still available to faculty

September 19, 2011

“Contagion” is just a hit movie but its premise – a deadly disease spread around the world by airline passengers – is no fiction, as recent experience with SARS and the H1N1 flu virus illustrate.

That’s one reason mechanical engineering Professor Qingyan Chen and his students are looking at aircraft environmental control systems with an eye to making them less apt to circulate airborne contaminants, as well making the systems less of a drain on fuel consumption.

Chen, principal investigator for the Air Transportation Center of Excellence for Airline Cabin Environment at Purdue, combines physical experiments and intensive computer modeling in his lab’s research – soon to be aided by Purdue’s new Hansen cluster supercomputer.

“Our research involves the solution of very complex equations and then we do iterations, which is why it takes so much computing power,” says Chen, whose models not only consider the three-dimensional nature of air circulation in spaces like airline cabins but often a fourth dimension, changes over time.

ITaP staff and student employees built Hansen, part of Purdue’s award-winning Community Cluster Program, in August and September. Besides Chen, faculty and students in earth and atmospheric sciences, chemistry, physics, computer science, aeronautics and astronautics, electrical and computer engineering, materials engineering, and more will use the new supercomputer.

Materials scientist Marisol Koslowski will use Hansen for large-scale simulations of deformation in nanocrystalline materials, many with uses in the energy industry. Among other things, the goal is to better understand factors leading to materials failure.

“We want to understand things at the microscale,” says Koslowski, a mechanical engineering professor. “But to understand things at the microscale you have to model at the nanometer scale. Your resolutions have to be very fine and that's why you need big machines.”

Room for expansion remains in the new cluster, named for the late Arthur Hansen, as Purdue president a strong supporter of high-performance computing at the University.

The Hansen cluster features Dell compute nodes with four 12-core AMD Opteron 6176 processors (48 cores per node). The nodes are available with either 96 or 192 gigabytes of memory. A large-memory option with 512 gigabytes of memory also is available. In addition, the new cluster features a high-performance Lustre scratch storage system and 10 gigabit Ethernet interconnects.

With Hansen, ITaP also is introducing an online application that lets faculty researchers directly manage the list of students and collaborators authorized to access their cluster capacity and run research jobs. Adding a user is as simple as a few keystrokes.

Through community clustering, faculty partners and ITaP pool funds to make more computing power available for Purdue research projects than faculty and campus units could afford individually.

ITaP’s Rosen Center for Advanced Computing installs, administers and maintains the community clusters, including security, software installation and user support, so researchers can concentrate on doing research rather than on running a high-performance computing system.

“I know if something goes wrong someone will be there to fix the problem,” Chen says “I’m a mechanical engineer. I’m not really a computer engineer or a computer scientist. It’s better to have a professional do it.”

Community clustering also maximizes the use of resources by sharing computing power among the faculty partners whenever it is idle. Researchers always have ready access to the capacity they purchase, and potentially more if they need it.

The program also is an opportunity for departments or individuals purchasing computing equipment for use outside the new cluster to take advantage of group pricing. Vendors provide significant price breaks on large orders. Purdue saved nearly $2 million versus standard university pricing on the Steele, Coates and Rossmann clusters built in 2008, 2009 and 2010.

The three clusters have delivered more than 300 million research computing hours to faculty and their students. With Hansen, that figure should hit half a billion hours later this year.

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167, gkline@purdue.edu

Originally posted: September 19, 2011