Abstract :
The efficiency of peer-to-peer (P2P) systems is largely dependent on the overlay constructions. Due to the random selection of logical neighbors, there is serious topology mismatch problem between the overlay and the physical topologies in many P2P systems. Such mismatching causes unnecessary query message duplications on both overlay and IP level, as well as increase query response time. In this research, we define the optimal overlay problem, and prove its NP-hardness. To address this issue, we propose a distributed overlay optimization algorithm, THANCS, and evaluate its effectiveness through trace driven simulations. The proposed THANCS has four major strengths. First, it does not need any global knowledge. Second, its optimization convergent speed is fast. Third, it is orthogonal to other types of advanced search approaches. Fourth, it reduces both the traffic cost and the search latency.
