My name is Joshua Breslau. I am a Princeton University graduate student, and am interested in attending the ACTS workshop in September. My advisor, Steve Jardin, has already submitted his recommendation.

I am in the plasma physics division of Princeton's astrophysics department, and have conducted my research at the Princeton Plasma Physics Laboratory. My thesis research has been focused on a numerical study of the basic plasma physics phenomenon of magnetic reconnection. I have written a new parallel, 2D resistive magnetohydrodynamic/two-fluid code, and have used it to study reconnection in a variety of parameter ranges. Variable mesh spacing, an implicit numerical scheme, and parallelization have made it possible for the first time to solve the global reconnection problem while still resolving the narrow reconnection layer adequately.

The code was developed on the T3E at NERSC and concurrently on an Origin 2000 at Princeton. It scales well up to about 24 processors, and has been successfully ported to the IBM SP and to a cluster of Pentium PCs as well. A typical run takes on the order of 100 CPU hours on the faster machines. For the next phase, I would like to extend my code to three dimensions, probably through the use of Fourier series. I believe that the ACTS toolkit would be very helpful in this effort.