摘要
节点选择算法是影响P2P系统带宽利用率和吞吐量的关键技术之一。P2P应用存在逻辑路径和物理路径之间不一致、忽略了覆盖网拓扑与底层网络拓扑之间的关系等问题。邻居节点间上传和下载能力、稳定性会影响传输速率。针对这一问题提出基于统计学习的方法构建邻居网络,同时优先选择上传能力强、稳定性好的邻居节点。计算机仿真实验表明,新算法能显著提高P2P系统的整体吞吐量,减少用户的平均下载时间,从而有效地改善P2P系统的整体性能。
Node selection algorithm is one of the key technologies which affect the P2P system bandwidth utilization and throughput. Neighbor node selection algorithm based on IP address information library provided by the network operator （ISP）, the location information is inaccurate or not timely updated. Neighbor upload and download capabilities, and stability will affect the transmission rate. In this paper, the method based on statistical analysis was proposed to build neighbor network, while giving priority to select the neighbors with good upload ability and stability. The computer simulation results show that the new algorithm significantly improves the overall throughput of the P2P system, reduces the average download time for the user, thus effectively improving the overall performance of P2P systems.
出处
《计算机应用》
CSCD
北大核心
2013年第A01期8-10,共3页
journal of Computer Applications
关键词
点对点
节点选择算法
统计学习
吞吐量
P2P
node selection algorithm
statistical analysis
throughput
