Advanced Computing, Mathematics and Data
Research Highlights

August 2008

Annual DOE Summer School Makes Computing Fun for Students

"I'm having fun here," said Gbenga Oyedepo, one of 20 students attending the 2008 DOE Summer School in Multiscale Mathematics and High Performance Computing. This year's workshop was held on the campus of Washington State University Tri-Cities and broadcast over the internet.

Oyedepo, a graduate student from the University of North Texas, particularly enjoyed the way that one of the speakers presented his talk on an area that he didn't intend to study, but "it got me thinking and makes me want to do more research in that area," he said. Sponsored by DOE's Office of Advanced Scientific Computing Research and organized by the Northwest Consortium of Multiscale Mathematics and Applications, the Summer School is helping to spur continuing advances in multiscale mathematics and high-performance computing, areas that are critical to helping solve complex problems in environment, energy, and national security.

The annual workshop seeks to develop and equip early career researchers in advancing the understanding of the mathematics of complex physical processes that occur on a wide range of interacting length and time scales. Aimed at graduate and post-doctoral students in mathematics, scientific computing, materials sciences, geophysics, computational physics, and mechanical engineering, this year’s event focused on research talks given by faculty from PNNL, Washington State University, Oregon State University, Princeton University, Interactive Supercomputing, and Galois, Inc. The talks included:

• Evolution of Computational Approaches in NWChem
• Multithreaded Systems for Emerging High-Performance Applications
• Overview of the Global Arrays Parallel Software Development Toolkit
• Specialising Generators for High-Performance Monte-Carlo Simulation ... in Haskell
• Knowledge Discovery with Star-P
• Hybrid Particle Simulations: An approach to the spatial multiscale problem
• Coarse-grained projective schemes for particle filters.
• Graph-based Pattern Learning
• Homogenization of Multiscale Problems
• Computational multiscale methods for coupled phenomena in subsurface

"The summer school is invaluable in exposing students to current and advanced topics in research done at PNNL, such as the leading-edge multiscale mathematics and multithreading systems for high-performance computing work our researchers are conducting," says PNNL researcher Daniel Chavarria, an instructor for this year's event. "I would have liked to attend a school like this when I was a student."

This is the third year for the Summer School. The first workshop was held in 2006 in Tacoma, WA, followed by a second Summer School at Oregon State University in Corvallis, OR in 2007.