DOE ACTS COLLECTION Workshop Robust and High Performance Tools for Scientific Computing September 4-7, 2002

(see recommendation letter)

I am a PHD student of USC. I am working with Professor Tom Katsouleas on simulation of plasma accelerator. We use a PIC code osiris to study the nature of wake field excited by electron beam or laser in plasma. We are planning to add new physics to the code about ionization of neutral gas. Also we plan to work on large-scale computation in the future. Recently, we did one run on SP nersc with 64 nodes; it ran for almost 2 weeks. So we are eager to look for a way to make our program faster, and more convenient interface to check the run or result.

(from Prof. T. Katsouleas): Suzhi has been working for me for with half time support from the SciDAC grant on Accelerators for approximately one year. In that time she has made significant contributions to the parallel PIC code OSIRIS supported by UCLA and USC. She has designed and implemented a new module to handle impact ionization in a gas/plasma in 2-D cylindrical and 3-D geometry. The module is already being used to design the next Plasma Wakefield Accelerator experiment at SLAC in 2003.

In my opinion, Suzhi is at the optimal point in her academic development to take advantage of this workshop. She is familiar with high performance computing and motivated to learn more. I expect what she learns from the workshop she will bring back and share with the group at USC and at UCLA invovled in high performance computing.