Elizabeth R. Jessup

Ours is a undergraduate course in high performance scientific computing. The course is designed to teach undergraduate students how to use high-performance computing systems in scientific and engineering applications and, in this way, to prepare them for work in the field of computational science. This course has been taught at UCB since 1991. The course covers two semester, but it's first semester stands alone to accomodate students who don't have time in their schedules for the full sequence. This course development has been supported by the National Science Foundation under a CISE Educational Infrastructure grant, CDA-9017953 since 1990. As far as we know, this is the first course of its type.

The course begins with introductions to the machines and performance measurement. Students also study the computational and visualization tools Matlab and AVS. The concepts and tools learned in the early chapters are then applied to solution of numerical problems in molecular dynamics, advection, and tomography. Because no available text covers all of these topics, we have written all of the course materials. These include a series of single-topic tutorials, a laboratory manual, short reference guides, and manuals for the machines and computational tools. Students use DEC and SGI workstations and a iPSC/2 hypercube multiprocessor located on the UCB campus as well as a Cray Y-MP (at NCAR) and a CM2 (at NCSA). Students access the latter two machines via the internet.

The course meets for two 50-minute lectures and one three-hour supervised laboratory period per week. Students work through a chapter of the lab manual every week. This involves understanding some complete example problems then extending the concepts learned by working through pencil-and-paper, programming, or visualization exercises.

In its first three years, our course has attracted juniors, seniors, and first year graduate students majoring in computer science, physics, applied mathematics, and mechanical or aerospace engineering. The only prerequisite is one semester of numerical computation at the undergraduate level.

Our course materials are being published by MIT Press and will appear as a textbook this summer. We are presently preparing a publishable version of the laboratory manual. Our materials have been available by anonymous ftp, and courses based on those materials are being offered at other institutions, both research universities and smaller 4-year colleges.