Rg. Hohlfeld et al., IMPLEMENTATION OF PARTICLE-IN-CELL STELLAR DYNAMICS CODES ON THE CONNECTION MACHINE-2, Journal of supercomputing, 7(4), 1993, pp. 417-436
The development of massively parallel supercomputers provides a unique
opportunity to advance the state of the art in N-body simulations. Th
ese N-body codes are of great importance for simulations in stellar dy
namics and plasma physics. For systems with long-range forces, such as
gravity or electromagnetic forces, it is important to increase the nu
mber of particles to N greater-than-or-equal-to 107 particles. Signifi
cantly improved modeling of N-body systems can be expected by increasi
ng N, arising from a more realistic representation of physical transpo
rt processes involving particle diffusion and energy and momentum tran
sport. In addition, it will be possible to guarantee that physically s
ignificant portions of complex physical systems, such as Lindblad reso
nances of galaxies or current sheets in magnetospheres, will have an a
dequate population of particles for a realistic simulation. Particle-m
esh (PM) and particle-particle particle-mesh (P3M) algorithms present
the best prospects for the simulation of large-scale N-body systems. A
s an example we present a two-dimensional PM simulation of a disk gala
xy that we have developed on the Connection Machine-2, a massively par
allel boolean hypercube supercomputer. The code is scalable to any CM-
2 configuration available and, on the largest configuration, simulatio
ns with N = 128 M = 2(27) particles are possible in reasonable run tim
es.