Application of Distributed and Parallel Computing in Traffic Network Simulation
Traffic network simulation describes movement behaviors and track of individual vehicles. When network is scale-up,the process takes more time and computing resources.Moreover, computing capacity of single processor computer cannot satisfy simulation requirement. A distributed and parallel computing approach to such a large-scale traffic system might be economical and efficient. This paper designs framework of parallel traffic simulation system by using domain decomposition method according to parallel characteristics of traffic network simulation, in other words,traffic network is partitioned to several sub-networks, and each processor of clusters is responsible for a different sub-network of the simulated region. A network-partition algorithm based on the number of vehicle is used to optimize load balancing. Information exchange mechanism between sub-networks is described. The developed parallel simulation system is implemented in parallel computing platform based on Message Passing Interface. From the results of numerical example, the proposed parallel simulation algorithm can improve speed and efficiency of traffic network simulation. This algorithm provides a foundation for larger scale traffic network simulation and real-time applications.
Traffic Network Simulation Distributed and Parallel Computing Domain Decomposition Network-Partition Algorithm
Zhicai Juan Linjie Gao Hongfei Jia
Institute of Transportation Studies, Shanghai Jiao Tong University Shanghai, China Transportation College, Jilin University Changchun, Jilin, China
国际会议
杭州
英文
108-112
2006-10-12（万方平台首次上网日期，不代表论文的发表时间）