ContextPeers: Scalable Peer-to-Peer Search for Context Information

Tao Gu,Hung Keng Pung,Edmond Tan,Daqing Zhang

2005-01-01

Abstract:Context information has emerged as an important resource to enable autonomy and flexibility of applications. The widespread use of context data necessitates efficient search services in wide- area networks. Centralized context search engines exist, but they suffer from single point of failure and single processing bottleneck problems. In this paper, we propose an efficient distributed context search system based on multiple overlay networks. Nodes that provide similar context data are semantically grouped into an unstructured Peer-to-Peer overlay. Context search requests are only routed to the appropriate overlays, reducing unnecessary query traffic on nodes that have irrelevant context data, and increasing the chances that context data will be found quickly. As the blind flooding mechanism generates a large volume of redundant messages, it will make the system unscalable. To reduce redundant query messages, we propose a Cost-Aware Selective Flooding technique. This technique makes use of 2-hop neighborhood and link cost information to ensure that only necessary messages are being flooded across the network. Our simulation studies demonstrate that our system is scalable and the intra-overlay and inter-overlay routing techniques are effective.