大规模网络拓扑探测和路径性能测量方法研究

A Study of Approaches of Topology Probe and Path Properties measurement in Large- Scale Network

网络拓扑探测和路径性能测量是大规模网络性能测量中的重要部分,是优化网络结构、提升网络性能的基础;对大规模网络环境下的拓扑探测和路径性能测量的关键技术进行了研究,针对传统Doubletree网络拓扑探测算法的不足,提出了一种改进后的算法Boftree,经过实验分析,在探测到相同节点数量的前提下,新算法减少了路径探测次数和探测过程中产生的网络传输开销;并介绍了一种基于矩阵理论的网络路径性能准确重建方法,该方法把网络路径中的链路性能参数抽象为矩阵元素,通过求解矩阵方程,得到网络中所有路径的性能参数值。

The network topology probe and path property measurement are the key components of the measurement of large--scale network per- formance, also fundamental to the optimization of network structure and properties. In this paper, we do research on several key technologies--of the topology probe and the path properties measurement in tile large--scale network environment. Furthermore, an improved algorithm named Boftree is proposed against the insufficiency of the traditional Doubletree topology probe algorithm. Based on experiments and result analysis, given the detection of the same number of nodes, Boftree algorithm needs less times of path probe as well as the overhead during the probing process. Finally, we intro- duce an accurate reconstruction method of network path properties based on the matrix theory. This method gets the value of each link parameter by solving a matrix equation whose elements are the link parameters to be measured.