𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Accelerating a Particle-in-Cell Simulation Using a Hybrid Counting Sort

✍ Scribed by K.J Bowers


Publisher
Elsevier Science
Year
2001
Tongue
English
Weight
503 KB
Volume
173
Category
Article
ISSN
0021-9991

No coin nor oath required. For personal study only.

✦ Synopsis


In this article, performance limitations of the particle advance in a particle-in-cell (PIC) simulation are discussed. It is shown that the memory subsystem and cachethrashing severely limit the speed of such simulations. Methods to implement a PIC simulation under such conditions are explored. An algorithm based on a counting sort is developed which effectively eliminates PIC simulation cache thrashing. Sustained performance gains of 40 to 70 percent are measured on commodity workstations for a minimal 2d2v electrostatic PIC simulation. More complete simulations are expected to have even better results as larger simulations are usually even more memory subsystem limited.


πŸ“œ SIMILAR VOLUMES


A rapid method for counting cell nuclei
✍ Andy A. Lin; Thang Nguyen; William M. Miller πŸ“‚ Article πŸ“… 1991 πŸ› Springer-Verlag 🌐 English βš– 187 KB

We report here an improved method for nuclei counting utilizing Triton-X 100 to reduce the size of cell debris, thereby allowing the use of a particle sizer]counter. Furthermore, nuclei are completely released within 30 seconds, as compared to 1 hour using hypotonic solution. The method is accurate

Orbital debris impact simulation using a
✍ Eric Fahrenthold; Ravishankar Shivarama πŸ“‚ Article πŸ“… 2001 πŸ› Elsevier Science 🌐 English βš– 756 KB

Simulations of three dimensional orbital debris impact problems, using a parallel hybrid particle-finite element code, show good agreement with experiment and good speedup in parallel computation. The simulations included single and multi-plate shields as well as aluminum and composite shielding mat