Join us at Super Computing 2011!We invite you to visit us at the SC2011 conference in Seattle, Nov 14-17 at Booth #2040 See first-hand how we are enabling research discovery with Dell HPC solutions.
Introduction by Dell's John FragallaWhen one thinks of a Supercomputer, one often thinks of large scale HPC system occupying a complete datacenter. This year at SC10 there will be a student challenge that will consist of a team to design, architect and run various HPC applications consuming a maximum power of no more than 26 amps. The architecture must utilize commercially available High Performance Computing (HPC) technologies.Dell will be sponsoring a cutting-edge, power efficient, HPC system to the University of Texas at Austin for this challenge, which will consist of the following team members: Bethany Barrientos, Vladimir Coxall, Alex Heinzmann, Jason Kilman, Loren Micheloni, and Phillip Verheyden. Their mentor is Byoung-Do Kim, a research associate in the High Performance Computing Group at the Texas Advanced Computing Center (TACC). The team of students will receive their Dell HPC system by the end of September, 2010, which will give them enough time to prepare for the competition at Super Computing 2010.
By: Laura Fidelman and Faith Singer-Villalobos, Texas Advanced Computing Center (TACC) Supercomputing 2010 (SC’10) in New Orleans will host the fourth Student Cluster Competition, challenging students to build, maintain, and run the most-cutting edge, commercially available HPC architectures on just 26 amps at 120 volts of electricity (or the power equivalent of three standard-size coffee-makers).The goal is to achieve the best cluster performance, with the most accurate outputs from application runs, while staying at or below the allotted energy budget. The teams hail from The University of Texas at Austin, Florida A&M University, Louisiana State University, the University of Colorado, Purdue University, Stony Brook University, National TsingHua University (Taiwan), and Nizhni Novgorod State University (Russia).The University of Texas at Austin’s team members—Bethany Barrientos, Vladimir Coxall, Alex Heinzmann, Jason Kilman, Loren Micheloni, and Phillip Verheyden—submitted a strong application outlining why they wanted to participate: “This project will provide us with the opportunity to gain an immense amount of experience in a research field that fuels our passion.”Their mentor is Byoung-Do Kim, a research associate in the High Performance Computing Group at the Texas Advanced Computing Center (TACC), a nationally renowned supercomputing center and a research unit of The University of Texas at Austin. Kim has worked at TACC since 2007, and has nine years of experience in the field. “I took the position as team supervisor because it sounded like a challenge,” Kim said. “Plus, working with these students is a lot of fun.”TACC secured their hardware sponsorship from Dell Inc. “Dell and TACC have shared a long and valuable partnership, so we are honored to support the TACC team for the SC10 Cluster Challenge,” said Tim Carroll, senior manager of Dell’s Research Computing Solutions group. “Bethany, Vladimir, Alex, Jason, Loren and Phillip have demonstrated true intellect, teamwork, and determination during their preparations for the Challenge, motivating us to support their efforts with our winning technology.We’re excited for the competition and will be cheering them on.”Kim says the precise configuration of the Dell hardware is top secret. “If I told you what it was, I’d have to kill you,” he says jokingly. The students have individual specializations in network topology, HPC visualization, low-level system design, and parallel programming; yet, their experience with HPC as a whole is limited. “I was aware of HPC because I worked at TACC, but I hadn't done any programming or anything related to HPC,” said Barrientos, a team member and former student intern with TACC’s Visualization and Data Analysis group.The team has a regimented practice schedule complimented by a staggering amount of independent research and testing. “We’re trying to learn as much as we can about HPC, as well as all the applications and the benchmarks, so we're ready to run everything when our Dell hardware comes in,” Micheloni said.