Research And Implementation Of Active And Passive Combined Network Topology Measurement Methods

With the increase of network scale and increasingly complicated network structure, it becomes more and more difficult for effective network topology measurement. However, network topology measurement is important to strengthen the network management, instruct the network design and improve the network performance. In this paper, we study the methods used by network topology measurement aim to improve the efficiency.The main contribution of this paper is as follows:We make a lot of research on the related technology and methods of network topology measurement. We conclude the limiting factor by analyzing the methods that network topology measurement takes from both active and passive measurement. It points the direction for searching more scientific and effective methods of network topology measurement.We propose an active network topology probe method based on binary search named BST and convert network pathset to network topology information.BST computes the hops of network paths and looks up the duplicate nodes probed. After that, it sends probe packets in parallel. BST increases the speed of topology and improves the efficiency significantly. We take network probe to deal with router alias resolution which brings out the topology information.We also present a passive topology information conversion algorithm based on OSPF named LCT. From automated acquisition of the network link state database using Expect, LCT algorithm converts it to network topology information. Dynamic monitoring is also presented for dealing with changes of network topology.We design and implement an active and passive combined network topology measurement system. Single method is hard to obtain good effect. Thus, active and passive combined method can get more complete network topology information. Under the guidance of passive method, probe the network by active.Network topology visualization system shows the topology by active and passive probing result.Our study provides ideas and show clearly direction for real-time, fast and accurate network topology measurement.