Competitive Mechanism Based Resource Distribution Strategy Research In P2P Streaming Media System

In recent years, streaming media technology with P2P as the core has beenwidely used. The status of node in the peer-to-peer network are equal, which hadno servers and clients. P2P technology has the features of efficient, fast and hidden, soall kinds of P2P networks has a huge user.Nodes in the P2P streaming media system should be fully mutual cooperation,sharing with the resources available, however, it is far from the real environment.Most users do not want to share their own resources, only a handful of selfless userscontribute resources and become service providers and response to the request of theother users. This phenomenon resulted to the sharing resources in the networkcontinue to decrease, and seriously affect the performance and fairness of the P2Pnetwork system. In order to solve the above problem, a resource allocation strategybased on the competitive is put forward. It not only can be achieved dynamicadjustment of the sending node to send media data, but also incentive each receivingnode to upload. So, it inhibit the free-rider, which maximize the benefits of a singlenode, also promote various types of nodes in the system to contribute actively, inorder to achieve maximum system fair and effective. The strategy consists of twoparts:The first part of the strategy to solve to define and calculate the value ofcontribution and benefits of each node. How to evaluate the contribution value of thenode, the paper mainly consider two factors, one is the number of nodes response andforward messages, and the other is the size of uploaded and downloaded fileof nodes. The benefits of nodes mainly consider the size of resources to get.The second part of the strategy to solve how to provide different services. Thispaper proposed the competitive model RDS-C and use the dynamic programmingalgorithm to allocate bandwidth for each competition node. Nodes with highcontribution value provide better services and maximize use of limited bandwidth,guaranteed bandwidth allocation fairness. In addition, the method also played apurpose to incentive selfish node to upload bandwidth to get more benefits.NS2simulation results show that the RDS-C has good incentive properties in thesame download bandwidth. Nodes with higher contribution value share morebandwidth, and nodes with lower contribution value share less bandwidth. Thus, theP2P network system provided different service to the node, each node in order to getmore benefits from the entire P2P network system had to provide more resources toother nodes, inhibition of selfish nodes in the network, inspire each node to contributeto the entire P2P network, greatly improving the efficiency of the system as a whole.