论文部分内容阅读
人工交通系统概念的提出为智能交通系统研究开辟了新的视野,但其自底向上的交通个体建模方法和复杂的计算过程也带来了庞大的计算量与存储量,单台计算机的计算能力越来越难以满足大规模计算实验的需求。为此,提出和实现了一种混合P2P结构的分布式计算实验框架,并采用改进的同步算法和新的负载均衡方案优化了框架的性能,最后通过测试验证了该框架的高效性和可靠性。
The concept of artificial traffic system has opened up a new field of vision for ITS research, but its bottom-up traffic individual modeling method and complicated calculation process also bring a huge amount of computation and storage. The calculation of a single computer Capabilities are increasingly difficult to meet the needs of large-scale computational experiments. To this end, a hybrid experimental framework for distributed computing with P2P structure is proposed and implemented, and the performance of the framework is optimized by adopting an improved synchronization algorithm and a new load balancing scheme. Finally, the validity and reliability of the framework are verified by tests .