Multi-objective Peer-to-Peer Neighbor-Selection Strategy Using Genetic Algorithm

Peer-to-peer (P2P) topology has significant influence on the performance, search efficiency and functionality, and scalability of the application. In this paper, we present a Genetic Agorithm (GA) approach to the problem of multi-objective Neighbor Selection (NS) in P2P Networks. The encoding representation is from the upper half of the peer-connection matrix through the undirected graph, which reduces the search space dimension. Experiment results indicate that GA usually could obtain better results than Particle Swarm Optimization (PSO).

• Type: Conference
• Year: 2007
• Where: HIPC
• Authors: Ajith Abraham, Benxian Yue, Chenjing Xian, Hongbo Liu, Millie Pant
• Keywords: Distributed And Parallel Computing, HIPC 2007, Multi-objective Neighbor Selection, Particle Swarm Optimization, Search Space Dimension