Research On The Performance Evaluation And Pricing Strategy For The Online Mechanism Of P2P Nodes

In recent years, with the rapid development of the P2P technology, P2P network hasbecome one of the key branches of Internet. How to regulate the selfish behavior of P2Pnodes and improve network resource utilization as well as service performance is a hottopic of research. Based on the online mechanism of the P2P nodes, continuous-timeMarkov models are constructed, and system performance is evaluated. A pricing policy isproposed with game theory, the P2P network is optimized socially.Firstly, according to the online mechanism of mobile P2P nodes, considering thecharacteristics of the strong mobility and the online/offline at any time, a two-dimensionalcontinuous-time Markov model with random number of servers is built. By constructing atransition rate matrix, the system model is analyzed in steady state with the method ofmatrix geometric solution. The definitions and expressions of the performance measures,such as failure rate and average delay of mobile P2P nodes are given.Secondly, based on the online mechanism of hybrid P2P nodes, the node state isclassified as online and offline. Taking into account the number of request nodes andonline hybrid P2P nodes, a three-dimensional continuous-time Markov model withrandom number of servers is established. By constructing another transition rate matrix,and using the matrix geometric solution method, the system model is analyzed in steadystate. The expressions of failure rate and average delay for hybrid P2P nodes are obtained.Then, for the mobile P2P network and hybrid P2P network, numerical experimentswith analysis and simulation are provided with different arrival rate and leaving rate of therequesting nodes and the service nodes. The system performance is evaluated bydemonstrating the change trends of these performance measures.Finally, considering transmission rewards and sojourn costs, net benefit functions areestablished. Nash equilibrium and the socially optimization for both the mobile P2Pnetwork and the hybrid P2P network are investigated. Numerical results show that thenode arrival rate with Nash equilibrium strategy is higher than that with socially optimalstrategy. A reasonable pricing policy is proposed for the requesting nodes, then, the P2P network is optimized socially.