冗余最小化的IPv6拓扑发现方法

Topology Discovery with Smallest-Redundancy in IPv6

随着网络技术的高速发展,网络管理的重要性越来越突出,正确的网络拓扑是进行网络管理的基础.IPv6是公认的下一代互联网协议,其庞大的地址空间和独特的特征为拓扑发现带来了新的挑战.目前,基于ICMP的拓扑发现分为分布式和集中式两种,其主动探测的特征不可避免地产生探测冗余.分布式拓扑发现方法布署困难并且成本高.更重要的是在冗余减少上存在由探测点间冗余引起的诸多限制,因此它不能以网络友好的方式发现拓扑.由于IPv6路由器对源路由的支持,集中式的拓扑发现方法能够发现交叉链路以保证覆盖率.测量了IPv6环境下单个探测源产生的冗余,提出了冗余最小化的集中式拓扑发现方法.在引入减少冗余的后退算法基础上提出了实际网络环境下的改进算法,说明了集中式拓扑发现在IPv6环境下的可行性.实验结果表明对靠近探测点的节点减少了高达两个数量级的冗余,并能够保证令人满意的覆盖率.

The importance of network management is more and more apparent, along with the quick evolution of network technology. Correct network topology is the basis of network management. IPv6 has been recognized as the next generation Internet protocol. However, its large address space and special features bring new challenges for topology discovery. Nowadays, the topology discovery based on ICMP can be divided into distributed method and centralized method, both of which produce the probing redundancy inevitably for probing actively. The distributed topology discovery is difficult to deploy, and its cost is very high. It also has a lot of limitations in redundancy reduction, due to inter-redundancy. Thus, it can＇t discover topology in a network-friendly manner. Since routers in IPv6 support the source routing function, the centralized topology discovery method can cover the cross link. In this paper, the intra-redundancy generated by one monitor probing is measured, and then a centralized topology discovery method with smallest-redundancy is proposed in IPv6 environment. Based on the backward algorithm, an improved algorithm is presented to reduce redundancy in practical network environment. Meanwhile, the centralized topology discovery method based on source routing in IPv6 is proved feasible＇. The experiment results show that this method can reduce redundancy of nodes near the monitor by two orders of magnitude, and obtain satisfactory coverage as well.