DocumentCode :
510159
Title :
A Parallel Monte Carlo Simulation on Cluster System for Particle Transport
Author :
Dai, Rui-kai ; Li, Jian-xin ; Dong, Chun-li ; Li, Lei ; Yan, Bin
Author_Institution :
Nat. Digital Switching Syst. Eng. & Technol. Res. Center, Zhengzhou, China
Volume :
2
fYear :
2009
fDate :
7-8 Nov. 2009
Firstpage :
31
Lastpage :
35
Abstract :
In Monte Carlo simulations, the more particles you trace, the more accurate results you get. While recent improvements in hardware and software have feasible on personal computers, it can still take hours, even days to achieve a statistically meaningful result. Geant4 is a kind of Monte Carlo simulation tool solving the problem in nuclear physics. It can be used to accurately simulate the passage of particles through matter. It has been reported that Geant4 parallelization was successfully achieved on PC cluster. Problems are how it works on cluster system which is different from PC cluster. Since cluster system has much higher data transmission speed, the limited exchange speed of data in the PC cluster is gone. In this paper, two usual parallel modes were compared in parallelizing the Monte Carlo simulation. They were static scheduling mode and master-slave mode. Simulations show that parallel Monte Carlo simulation on cluster system for particle transport using Geant4 is feasible.
Keywords :
Monte Carlo methods; parallel processing; physics computing; Geant4; cluster system; master-slave mode; nuclear physics; parallel Monte Carlo simulation; particle transport; static scheduling mode; Artificial intelligence; Bandwidth; Computational intelligence; Computational modeling; Computer networks; Costs; Discrete event simulation; Hardware; Master-slave; Processor scheduling; Cluster system; Geant4; Monte Carlo Simulation; particle tranport;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Artificial Intelligence and Computational Intelligence, 2009. AICI '09. International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-3835-8
Electronic_ISBN :
978-0-7695-3816-7
Type :
conf
DOI :
10.1109/AICI.2009.485
Filename :
5376368
Link To Document :
بازگشت