Affiliations: School of Economics and Management, Yanshan University, Qinhuangdao, China
Correspondence:
[*]
Corresponding author. Wei Sun, School of Economics and Management, Yanshan University, Qinhuangdao 066004, China. E-mail: wsun@ysu.edu.cn.
Abstract: In the last years peer-to-peer (P2P) networks have become an important and prevalent architecture in the Internet and have been commonly applied into many scenarios such as big data storage and management, cloud computing, vehicular networks and social networking. The overall performance of P2P networks mainly depends on cooperation between peers, so as to encourage each peer to contribute its resource to others and reduce free-riding behaviors. Then each peer has to face a dilemma decision that how much of its available bandwidth it will allocate to uploads for others and how much it will reserve for its own downloads. In this paper we consider bandwidth allocation in P2P networks where both upstream and downstream flows traverse one common access link with finite capacity, and formulate the utility maximization model for bandwidth allocation. The model is difficult to resolve since it is an interplay problem between upload and download decisions of each peer. In order to achieve the optimal bandwidth allocation, we present a heuristic scheme using Particle Swarm Optimization (PSO), and discuss the performance with numerical examples in different network scenarios. Simulation results show that the scheme can achieve the global optimum within reasonable iterations.
