Distributed cache indexing for efficient subspace skyline computation in P2P networks

Skyline queries play an important role in applications such as multi-criteria decision making and user preference systems. Recently, more attention has been paid to the problem of efficient skyline computation in the P2P systems. Due to the high distribution of the P2P networks, the skyline computation incurs too many intermediate results transferred between peers, which consumes mass of the network bandwidth. Additionally, a large number of peers are involved in the skyline computation, which introduces both heavy communication cost and computational overhead. In this paper, we propose a novel Distributed Caching Mechanism (DCM) to efficiently improve the performance of the skyline calculation in the structured P2P networks, using a Distributed Caching Index (DCI) scheme and an advanced cache utilization strategy. The DCI scheme is employed to efficiently locate the cache that can properly answer a future skyline query. Exploring the property of entended skyline, we can optimize the utilization of the cached results for answering future skyline queries. Extensive evaluations on both synthetic and real dataset show that our approach can significantly reduce both bandwidth consumption and communication cost, and greatly shorten the response time