论文部分内容阅读
为解决P2P流媒体系统构建时的邻居选择问题从而优化应用性能,本文首先提出了一种新的评价模型-带宽延迟比率,然后基于该性能模型提出了一种基于Metropolis-Hastings算法同时考虑P2P流媒体系统中节点局部性和异构性的拓扑构建算法,并对提出的算法建立数学模型进行了理论的分析。最后本文对该算法进行了模拟实验,结果表明该算法与目前广泛采用的SCAMP方法相比具有显著的性能提高和更好的可扩展性。本文的研究工作对P2P流媒体系统的拓扑构建,性能优化等领域的研究有着重要的指导意义。
In order to solve the problem of neighbor selection in P2P streaming media system construction and optimize application performance, this paper first proposes a new evaluation model - bandwidth delay ratio, and then proposes a Metropolis-Hastings algorithm based on the performance model to consider P2P streaming Topological construction algorithm of nodes’ localization and heterogeneity in media system, and analyzes the proposed algorithm by establishing mathematical model. Finally, the algorithm is simulated in this paper. The results show that the proposed algorithm has significant performance improvement and scalability compared with the widely used SCAMP method. The research work in this paper has important guiding significance for the research of topological construction and performance optimization of P2P streaming media system.