Deployed in October 2009 by SDSC, an organized research unit of UC San Diego, the Triton Resource is a medium-scale "cluster" computing system with a respectable 29 Teraflops of peak computing power. Originally conceived as providing UC researchers with a significant SDSC-owned capability for conducting "data-intensive" science, the Triton Resource has since attracted the interest of professors conducting courses and workshops in several scholarly disciplines.
To date, courses involving the Triton Resource have included parallel programming, advanced computer architecture, astrophysics and cosmological simulations, and economics. Computer science professor John Gilbert from UC Santa Barbara is using the Triton Resource in an undergraduate course in parallel scientific computing (CS140) being offered for the Winter 2011 quarter, and a graduate course in applied parallel computing (CS 240A) to be offered this spring. CS 240A is required for UCSB's interdisciplinary graduate emphasis in Computational Science and Engineering.
"The Triton Resource plays an important role at UCSB in computer science and computational science education," said Gilbert, who has taught such courses for several years, always using at least one SDSC supercomputer. "High-performance computation is becoming crucial to more and more areas of science and engineering, and access to significant computational resources such as the Triton Resource is essential for training both undergraduate computer scientists and graduates."
"The interest in the Triton Resource as a teaching tool shows the increasing pervasiveness of computational science across many scholarly disciplines and the recognition that we need to equip our graduates with the ability to use high-performance computing in their academic and industrial careers," said Michael Norman, SDSC's director and an instructor for a cosmological simulation course offered in Summer 2010.
The Triton Resource has been well received by students taking a course in graduate level parallel computation, said Scott Baden, a professor in UC San Diego's Department of Computer Science and Engineering. Baden taught the first course using the Triton Resource in the Fall 2009 quarter, a course in parallel computation (CSE 260).
"We have seen a steady increase in enrollment and interest since this program began," Baden noted. "The Triton Resource rose to the challenge, providing an environment that encouraged experimentation and innovation."
"It's exciting to see that SDSC's initial investment, which focused on creating a research computing system, is also having an important impact on the University's teaching and training mission," added Philip Papadopoulos, Director of UC Systems at SDSC. "Students in these classes are being exposed to significant, state-of-the-art computing and that experience will make them more competitive when they leave UC for the next stage in their careers."
SDSC's Triton Resource was conceived and constructed by SDSC to provide a medium-scale, general purpose, and energy-efficient computing system. In its first year, the Triton Resource has supported research projects in many disciplines, including drug discovery and design, economics, genomics and metagenomics, microprocessor design, nanoengineering of solar cells, phylogenetics, and weather forecasting. Please click here for more information on the Triton Resource.
As an organized research unit of UC San Diego, SDSC is a national leader in creating and providing cyberinfrastructure for data-intensive research, and celebrated its 25th anniversary in late 2010 as one of the National Science Foundation's first supercomputer centers. Cyberinfrastructure refers to an accessible and integrated network of computer-based resources and expertise, focused on accelerating scientific inquiry and discovery. SDSC is a founding member of TeraGrid, the nation's largest open-access scientific discovery infrastructure.
Jan Zverina, SDSC Communications
858 534-5111 or email@example.com
Warren R. Froelich, SDSC Communications
858 822-3622 or firstname.lastname@example.org