Power-law Characteristic based Research on Unstructured P2P Broadcast Mechanism

摘要

The information source of spatial data is wide and the data quantity is huge. It is not suitable to manage and share spatial data under centralized structure. Because Peer-to-Peer (P2P) systems have shown great potential on file sharing, using P2P system to manage distributed spatial data is an efficient way. Flooding-based broadcasting is the widely used mechanism in many current large-scale P2P systems. This mechanism usually lead to serious communication cost problem. Based on the observation of the small world and power-low properties of unstructured peer-to-peer networks as large-scale, self-configure, and dynamic complex systems, a probabilistic model of message forwarding is presented. This model use rumor spreading mechanism and preferential spreading to accommodate the clustering and power law characteristics. Mathematical analyses and preliminary simulation show that this approach could sharply optimize the communication cost in current P2P systems based on flooding mechanism.