A hierarchical divisive algorithm for topology discovery in multi-subnet networks

Current progress in network techniques and the complexity of network structures require network administrators to spend more time and attention on maintaining a network system under normal operations. With our continuously increasing dependence on networks, it is difficult to know how to effectively manage all of the equipment in a network and to be certain of whether the network is operating correctly. To make a comprehensive survey of current administrators' management modes, administrators must understand the rules for the network's equipment and the entire topological structure of the network; such an understanding will allow them to rapidly and efficiently address an increasingly complicated network environment. This paper presents a hierarchical divisive algorithm for topology discovery that can work on a network layer and a link layer, which are widely used by network administrators today. This approach involves an automatic exploration through a random network node and possesses the features of cross-layer and multiple protocols to support the exploration of a multi-subnet network, to enable a faster and more effective explorative ability of a topological resource. The implementation shows how to automatically discover the entire network's topology and all of the network's equipment link nodes from the integration of a network-managed system to prove that our proposal is effective.