Purdue Team Driving SiCortex Low-power Supercomputer in 2008 Cluster Challenge

A team of student engineers at Purdue University in West Lafayette, IN will be using technology from SiCortex, a Massachusetts company that makes lower-power supercomputers, for its cluster computing entry in an upcoming competition. The Boilermaker Cluster Challenge team will be entering its design at SC08, a supercomputer conference put on in November by IEEE Computer Society and the Association for Computing Machinery.

The Cluster Challenge's rules limit the amount of power the entries can draw from two 120-volt, 13-amp circuits, which limits the number of processors that can be employed. Purdue's cluster had more than 40 processors last year, with most competitors falling between 30 and 60, said Jeffrey Evans, an assistant professor in the Department of Electrical and Computer Engineering Technology. The 2008 team may be able to run nearly 1,000 processing cores concurrently in the SiCortex.

Evans said working with so many processors promises to be a good experience for the students on the team. The students work on their cluster as part of his high performance computing class, normally graduate level but open to undergraduates on the team.

The 2007 Purdue team had to assemble its cluster from standard rack-mounted machines. Since the SiCortex comes assembled out of the box, all the hardware in one package and basic software installed, that should give the '08 team more time to install, configure and refine the benchmarking programs and working applications Cluster Challenge teams must run for the competition, Evans said.

"Hardware issues could come into play if something fails, or something begins misbehaving," he said.

Team leader Preston Smith, a senior UNIX system administrator for Purdue's Rosen Center for Advanced Computing, said the SiCortex machine presents some extra software challenges, however. The system's processors are atypical for one thing, so applications need to be adjusted to run on it.

The programs also need to be tuned to take advantage of the large number of processors and fast interconnect fabric SiCortex employs to offset the raw speed disadvantage of its lower-power chips, which run at 500 megahertz.

Teams, limited to six undergraduate students, have to apply to participate in the Cluster Challenge. Last year, six teams participated. Evans said Purdue's team thought the unusual nature of the SiCortex computer might boost its case for participation this year.

According to the 2008 Cluster Challenge Web site, the winning team will be chosen based on workload accomplished, benchmark performance, and overall architectural design.

Last year's Purdue team received a lot of attention for an interactive comic book created to promote the team and its Cluster Challenge entry. This year, Information Technology at Purdue (ITaP) and Rosen Center staff members are creating a Web-based game, kind of a Sim cluster, where players choose from different components, including SiCortex nodes, and try to balance their configurations to best handle a series of virtual jobs. Completed jobs earn grant money, the better to buy upgrades. Machines that don't handle the load well go bust.

In real life, Purdue team members already have started working with likely competition applications. The Rosen Center, the high performance computing and discovery arm of ITaP, installed a SiCortex machine in June to test it from both performance and energy-consumption perspectives. The loaner is an SC1458, the next step down from the Rosen Center's SC5832. The numbers in both cases refer to the maximum number of processors the machines can contain.

About the Author

Dian Schaffhauser is a former senior contributing editor for 1105 Media's education publications THE Journal, Campus Technology and Spaces4Learning.

Featured

  • closeup of person wearing abstract smart glasses

    Google Unveils Android XR Smart Glasses, Powered by Gemini AI

    More than a decade after the commercial failure of Google Glass, Google is returning to the smart-glasses market, this time betting that advances in artificial intelligence, miniaturized hardware, and conversational computing can turn wearable devices into a mainstream platform.

  • circuit patterns

    Anthropic Launches Lower-Cost Claude Sonnet 5

    Anthropic has released Claude Sonnet 5, positioning the model as its most autonomous mid-tier offering to date and a lower-cost alternative to its flagship Opus 4.8 system. The company said the model can plan multi-step tasks, operate tools such as browsers and terminals, and complete agentic work at a level that previously required larger and more expensive models.

  • Student classroom scene with diverse learners attentively engaging in lecture, using laptops

    The AI Literacy Gap No One Expected

    While Gen Z may be advanced at generating quick outputs or using free LLMs for surface-level tasks, they need to develop critical thinking, communication, and analysis skills.

  • abstract cybersecurity data protection

    Rubrik Intros Google Workspace Data Protection

    Rubrik has announced the launch of Rubrik Data Protection for Google Workspace, a product the company said is designed to help enterprise customers protect data and restore operations across Google Workspace environments.