| 微观交通分布式并行仿真系统设计与效益分析 |
| |
| 引用本文： | 高林杰,隽志才,倪安宁.微观交通分布式并行仿真系统设计与效益分析[J].公路交通科技,2006,23(4):94-98. |
| |
| 作者姓名： | 高林杰 隽志才 倪安宁 |
| |
| 作者单位： | 1. 吉林大学,交通学院,吉林,长春,130022
2. 上海交通大学,交通运输研究所,上海,200052 |
| |
| 基金项目： | 国家自然科学基金资助项目(50378042、50338030)，吉林省科技发展资助项目(20030565) |
| |
| 摘 要： | 为了提高交通网络仿真的速度和效率,从挖掘交通网络仿真的并行性出发,设计了分布式并行仿真系统的结构和仿真算法,通过比较选择PVM(Parallel Virtual Machine)作为并行计算的编程环境,提出基于车辆数的网络分割算法和数据时钟同步的通讯方式,使并行仿真达到负载平衡。用程序执行时间、加速比、并行效率和可扩展性4个测度指标来评价算法的并行效益,并以24个交叉口组成的交通网络为例,应用设计的并行算法在不同处理器数目下进行仿真。试验结果显示并行仿真速度比串行仿真提高了3.5倍,因此,这种并行算法提高了交通网络仿真的速度和效率,为实现更大规模的交通仿真奠定了基础。
|
| 关 键 词： | 分布式仿真 并行效益 网络分割算法 负载平衡 |
| 文章编号： | 1002-0268(2006)04-0094-05 |
| 收稿时间： | 2006-07-29 |
| 修稿时间： | 2005年7月29日 |
Design and Benefit Analysis of Distributed and Parallel Microscopic Traffic Simulation |
| |
| GAO Lin-jie,JUAN Zhi-cai,NI An-ning.Design and Benefit Analysis of Distributed and Parallel Microscopic Traffic Simulation[J].Journal of Highway and Transportation Research and Development,2006,23(4):94-98. |
| |
| Authors: | GAO Lin-jie JUAN Zhi-cai NI An-ning |
| |
| Institution: | 1.Transportation College, Jilin University, Jilin Changchun 130022, China; 2.Institute of Transportation Studies, Shanghai Jiaotong University, Shanghai 200052, China |
| |
| Abstract: | The design of distributed parallel simulation structure and algorithm according to parallel characteristics of traffic network simulation were presented.PVM(Parallel Virtual Machine) serves as parallel computing platform.It puts forward network-partition algorithm based on number of vehicles and communication means of synchronization to realize load balancing.Parallel benefits of simulation algorithm are assessed by four measures including program run time,speedup,parallel efficiency and expandability.With parallel algorithm,traffic network with 24 intersections is modeled based on different number of processors.The results from the simulation demonstrate that simulation speed increases 3.5 times.Hence,parallel simulation algorithm improves speed and efficiency of traffic network simulation.It provides a foundation for larger scale traffic network simulation. |
| |
| Keywords: | Distributed simulation Parallel benefits Network-partition algorithm Load balancing |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
