A team of three computational and research scientists at NASA's Jet Propulsion Laboratory in Pasadena, CA, are the recipients of a prestigious award for a key paper detailing their new code enabling faster, more efficient parallel computing. Ping Wang, Daniel Katz and Yi Chao were awarded the $1,000 Best Paper prize from a field of 384 entries at SC '97: High Performance Networking and Computing, a supercomputing conference held November 17-21. The event was co-sponsored by the Institute of Electrical and Electronics Engineers and the Association for Computing Machinery.
The trio's paper, Optimization of a Parallel Ocean General Circulation Model, discusses their successful efforts to design and implement a well-optimized code to significantly improve computational performance and reduce the total research time to complete complex studies.
The team applied this code to a three-dimensional simulation, known as an ocean general circulation model (OGCM), of the entire Atlantic Ocean, a task requiring a large amount of memory and processing time. The general strategies employed to optimize the OGCM included memory optimization, effective uses of arithmetic pipelines and use of optimized libraries.
The paper is available in its entirety at the Web site for JPL's High Performance Computing Group: http://www-hpc.jpl.nasa.gov/
The Jet Propulsion Laboratory is a division of the California Institute of Technology, Pasadena, CA.
818-354-5011