Measuring and Characterizing Topologies of P2P Networks

Abstract

PDF HTML阅读 XML下载 导出引用 引用提醒 对等网络拓扑测量与特征分析 DOI: 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: Supported by the National Natural Science Foundation of China under Grant No.60403033 (国家自然科学基金) Measuring and Characterizing Topologies of P2P Networks Author: Affiliation: Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:测量分析对等网络(peer-to-peer networks)拓扑特征是解决P2P优化、网络监管等问题的基础.对等网络是一类大规模、自组织、并且高度动态的复杂网络系统,准确、完整地测量所有对等网络拓扑面临很大困难.研究对等网络的协议特点,分析特定P2P拓扑实例成为认识P2P拓扑特性的一种可选研究方案.以Gnutella网络为测量对象,定义了对等网络拓扑测量系统准确性、完整性的衡量指标,设计、实现了基于正反馈的分布式Gnutella拓扑爬行器——D-Crawler;分析了Gnutella网络拓扑图的度等级分布特征、度频率分布特征以及小世界特性.实验和分析结果表明,对等网络拓扑图属性特征与其使用的协议和客户端软件行为密切相关;Gnutella网络中不同层次的节点之间的拓扑关系表现出不同的特性:上层节点组成的子图具有度等级幂律特征,但在其度频率分布上却呈现出正态分布的特性;下层节点在度等级分布上的幂律特征表现不强烈,而在其度频率分布特征上具有明显的幂律特性.拟合结果表明:幂律能够较好地拟合度等级分布和下层节点度频率分布,然而对于上层节点度概率密度分布,Gaussian拟合效果最好.Gnutella网络具有小世界特性,即:较大的聚集系数和较小的特征路径长度,但它不是无尺度图,不符合BA(Barabási-Albert)生长模型,其发展遵循一种不同于BA模型的生长过程. Abstract:Measuring and characterizing the topological properties of peer-to-peer networks will benefit the optimization, management of the P2P systems. It seems infeasible to capture a complete and precise snapshot of P2P overlay networks due to the variety of P2P protocols and dynamics of the servents. Studying the details of P2P protocols and analyzing the specific P2P overlay network instance become an alternative method for this goal. In this paper, the measured Gnutella network topology is basically taken as an example. The architecture and performance of the distributed crawling system (called D-Crawler system) with feedback mechanism is presented. The properties of degree-rank distributions and degree-frequency distributions of the measured topology graphs are analyzed in detail. The small world characteristics for Gnutella network are discussed. The results show that the P2P overlay network topology characters are closely related to the P2P protocols and clients' behaviors. Gnutella network shows different characters in each tier. The top level graphs fit the power law in degree-rank distribution, but follow the Gaussian function in degree-frequency distribution. The bottom level graphs show weak power law in its degree-rank distribution, but are power law in its degree-frequency distribution. Fitting results indicate that power law could fit better for the degree-rank distribution and degree-frequency distribution of bottom level graphs, Gaussian could describe the degree-frequency distribution of the top level graphs. Gnutella overlay network has the small world property, but it is not the scale-free network, its topology may have developed over time following a different set of growth processes from those of the BA (Barabási-Albert) model. 参考文献 相似文献 引证文献