Projects

As a convergence point for engineering, mathematics and computer science with the domain sciences, IDRE will represent and support research at UCLA where high performance computation, visualization, modeling and simulation play a key role.

Every quarter IDRE will also solicit internal proposals for focused research projects in the area of high-performance computing. The purpose of this annual solicitation is to help faculty and research groups identify and solve their "bottlenecks" in high-performance computing. Successful proposals will have access to expertise and resources of IDRE research scientists and scholars and post-doctoral researchers.

2008

Evolution of our Solar System

Researching the Evolution of our Solar System Brad Hansen, a professor in astrophysics, researches white dwarfs, extrasolar planets, neutron stars and black holes. By studying other planetary systems he learns how our own solar system may have evolved. He plans to spend his summer, however, learning to parallelize code to take advantage of his Contributed cores within the new Hoffman2 Research Virtual Shared Cluster. With a grant from the Sloan Foundation and support from his graduate students, Steve Berukoff and Elliot Koch who were familiar with cluster technology, and IDRE professional staff, his 80 core cluster is now up and running in the new IDRE data center. Asked why he chose to become part of the shared cluster, professor Hansen said that finding the appropriate space for housing the cluster seemed a bit daunting. He talked to Warren Mori, Director of IDRE and a fellow physicist, and learned of the program.  He and his team are planning to use their full cluster by summer's end and then eventually run a really big job to demonstrate the potential of harnessing unused cores across the entire Hoffman2 cluster.

Exact Many-Electron Simulation Method for Bond-Breaking Reactions in Liquids

(recipient of the IDRE 2007 call for proposals)

Professor Benjamin J. Schwartz and William J. Glover, Dept. of Chemistry & Biochemistry. IDRE-ATS Research Scientist Prakashan Korambath is working with this group to solve eigenvalue problem for a Sparse Matrix using Implicitly Restarted Lancoz Method to exact many-electron simulation method for bond-breaking reactions in liquids.

High-Performance Image Processing for Cryo-Electron Microscopy

(recipient of the IDRE 2007 call for proposals)

Professor Z. Hong Zhou and Xhark Zhang , Electron Imaging Center for Nanomachines, CNSI and MIMG. IDRE-ATS Research Scientists, Scott Friedman and Shao-Ching Huang are working with this group to speed up the code by performing FFT on the GPU for fast image processing and reconstruction utilizing the GPU.