Computer simulation using particles
Computer simulation using particles
A parallel particle-in-cell model for the massively parallel processor
Journal of Parallel and Distributed Computing - Massively parallel computation
Vector models for data-parallel computing
Vector models for data-parallel computing
Implementations of randomized sorting on large parallel machines
SPAA '92 Proceedings of the fourth annual ACM symposium on Parallel algorithms and architectures
Implementation of particle-in-cell stellar dynamics codes on the Connection Machine-2
The Journal of Supercomputing
Plasma Physics Via Computer
Sorting networks and their applications
AFIPS '68 (Spring) Proceedings of the April 30--May 2, 1968, spring joint computer conference
| Hi-index | 0.00 |
The authors investigate the most efficient implementations of the charge (mass) assignment and force interpolation tasks of a particle-in-cell code on the SIMD architecture of the MasPar MP2. Three different approaches were tested. The first emphasized uniform computational (not necessarily communication) load balance and ease of programming. The second exploited the speed of the Xnet interprocessor communication network using a particle data migration strategy. The third used sorting and vector scan-add operations on the particle dataset to minimize the communication traffic required between the particle and mesh data structures. Algorithm efficiencies were measured as a function of the degree of spatial clustering of the particles, and as a function of the total number of particles. The sort/scan-add strategy gave the best performance for a broad range of degree of spatial clustering. It was only beaten by the migration strategy in the regime of weak clustering. Their results indicate how a hybrid algorithm combining the migration and sort/scan-add strategies can set an upper limit on the performance degradation associated with the spatial clustering of particles.