Study of Network Size Measurement Algorithm for P2P System

Peer-to-peer (P2P) networks have recently gained much attention owing to their selforganization characters under dynamic and decentralized environment. It is important for a P2P routing to keep track of the underlying network topology or at least an estimation of the number of nodes in such a network. However, this is quite difficult due to the frequent arrival and leave of nodes in peer-to-peer networks. In order to solve this problem, two network size measurement algorithms are proposed: binary tree and range searching to suit the network fluctuate. The advantages and disadvantages of the two size measurement algorithms are discussed. The results show that both algorithms can efficiently measure the size of P2P networks when topology changes. Simulation results show that both of the proposed approaches can give a good estimate of network size. In addition, comparison of performance indexes such as resilience to Failure, measurement error and workload shows that the binary tree algorithm can give a better estimate at the maintenance cost.