Implement the PIC-MCC simulation of ECR plasma source on PC cluster with DRBL, an effective approach to integrate PC cluster

Author

Shiau, J.H. ; Hu, Y. ; Lin, T.L. ; Huang, K.L.

Author_Institution

Dept. of Eng. & Syst. Sci., Nat. Tsing Hua Univ., Hsinchu, Taiwan

fYear

2005

fDate

28-30 May 2005

Firstpage

311

Lastpage

314

Abstract

A parallel 2d3v PIC-MCC (particle-in-cell plasma simulations and Monte Carlo collisions) code is developed with message passing language PVM to simulate the electron cyclotron resonance plasma source in this research. This code has been run on a PC cluster with a cluster system we developed called "diskless remote boot in Linux (DRBL)", which is a centralized boot and management environment. By using DRBL, it\´s easier for us to integrate some PCs to a cluster system. Once we have some computing power, with the grid technology, we can harvest more by sharing the computing resources with others. This is a new approach to harvest more computing power. With the PIC-MCC simulation of ECR plasma source, the results and benchmark for different cluster architectures arc given and discussed.

Keywords

Monte Carlo methods; cyclotron resonance; grid computing; physics computing; plasma simulation; workstation clusters; Linux; Monte Carlo collision; PC cluster; PIC-MCC simulation; PVM language; diskless cluster; diskless remote boot; electron cyclotron resonance plasma source; message passing language; parallel 2d3v PIC-MCC code; particle-in-cell plasma simulation; Cyclotrons; Electrons; Environmental management; Grid computing; Linux; Message passing; Monte Carlo methods; Plasma simulation; Plasma sources; Resonance; Diskless cluster; ECR plasma source; PC cluster; PIC-MCC method;

fLanguage

English

Publisher

ieee

Conference_Titel

Cellular Neural Networks and Their Applications, 2005 9th International Workshop on

Print_ISBN

0-7803-9185-3

Type

conf

DOI

10.1109/CNNA.2005.1543223

Filename

1543223