Research Of Queueing System Based On P2P Network Resource Search Mechanism

P2P network technology as a key issue in the field of computer network has been studied mainly from four aspects:search,storage,security and application.Among them,resource search has always been the focus of research,but there are still many unsolved problems.This paper mainly studies the resource search field of P2P network,relevant knowledge of queueing theory is applied.The queueing model is used to quantify the P2P network resource search mechanism,three queueing systems based on P2P network resource search mechanism are studied,and the performance of the system is analyzed.The paper mainly includes the following three parts:Firstly,aiming at the uncooperative selfish behavior of nodes in P2P network,a queuing model of differentiated services is established based on queueing theory.The request nodes with high degree of cooperation enjoy higher service rate,while the request nodes with selfish behavior enjoy lower query service rate.According to the relevant state of the system,a three-dimensional Markov chain is built,and the queueing model of synchronous work vacation of some nodes is analyzed.The expression of the main performance indexes of the system under steady state are obtained by using the quasi-birth-and-death process and the matrix-geometric solution,numerical experiments are used to study the influence of parameter changes on the main performance indexes of the system.The individual income function is constructed and the social optimal strategy is analyzed.Secondly,aiming at the selfish behavior of nodes in P2P network who are unwilling to participate in resource query,a new resource search mechanism is established based on queueing theory.The queueing model with preemption priority,secondary optional service and synchronous multiple working vacations is constructed,the average stay time,total energy consumption and other performance indexes of the system are obtained by using matrix-geometric solution method.Through numerical experiments,the impact of parameter changes on each performance index of the system is analyzed.Finally,the utility function of the system is constructed,the personal and social benefits of resource query nodes are discussed,and the effectiveness of the model has been verified.Finally,the N-strategy is introduced into the queueing model,and a resource search mechanism based on threshold priority is built.By distinguishing the quality of service of nodes,the nodes are encouraged to respond positively to the queries in the system.The matrix-geometric solution is used to analyze the system stability,and the expression of the system performance indexes are obtained.The influence of parameter changes on the system performance index is studied,and the personal income function and social income function are constructed,the equilibrium strategy of individual income is discussed through game theory analysis.Numerical experiments show that the priority strategy can make the node gain more personal benefits and guide it to become a node with good behavior.The social optimal strategy is analyzed,and the optimal arrival rate of resource search request node is obtained when the social benefit is maximum.