1998 Annual Report
Year in Review

Computer Science and Applied Mathematics

Illustrating the complexity of a typical next-generation experimental physics event, this image shows only 5% of a simulated gold-on-gold nuclear collision at the center of the STAR detector at Brookhaven's Relativistic Heavy Ion Collider. Finding better ways to store, move, access, and analyze huge datasets from experiments like this is one of the goals of NERSC's computer science research efforts.


To better meet our clients' current and future needs, NERSC stays at the forefront of high performance scientific computing by participating in a broad range of computer science and applied mathematics research. We participate in the development and testing of next-generation software and hardware, and many of our research staff members develop advanced numerical algorithms for better solutions to scientific problems.

There were some organizational changes in 1998 to improve the integration of our research efforts. Bob Lucas joined NERSC as head of the High Performance Computing Research Department, and that department welcomed two existing organizations into its ranks: the Center for Computational Sciences and Engineering, headed by John Bell, and the Applied Numerical Algorithms Group, headed by Phil Colella.

The Data Intensive Distributed Computing Group, under Brian Tierney, was merged into the Future Technologies/Software Tools Group, led by Bill Saphir. And to focus bioinformatics efforts at Berkeley Lab and support the DOE's Genome Program, staff from NERSC, the Life Sciences Division, and the Information and Computing Sciences Division joined to form the Center for Bioinformatics and Computational Genomics under the joint management of Sylvia Spengler and Manfred Zorn.

NERSC's management team was completed by the addition of Bob Lucas as head of research.

NERSC staff research efforts this year included bioinformatics, data-intensive storage access, parallel software libraries, distributed computing, fluid dynamics algorithms for combustion simulations, and cluster architectures. These projects are described below.


 INDEX  NEXT >>