Approach for service search and peer selection in P2P service overlays

Peer-to-peer service overlay networks (P2P SON) are increasingly being infrastructure providers for networking services, allowing service providers to cooperatively offer and run a flexible set of services. Regarding this condition, the selection of peers is a key issue for improving resource usage; service performance, and ultimately end users Quality of Experience (QoE). This paper presents an approach to best peer selection in a three-tier P2P SON architecture, allowing the splitting of service business functions and peer selection functions. The proposed best peer selection approach is evaluated by simulation, using a literature available geographic positioning metrich that takes into account real delay and jitter made available by the CAIDA project and MaxMind´s free database. The simulation results show the consistency and good performance of the proposed peer selection approach.