Abstract:

Search is one of the most important problems in peer-to-peer (P2P) networks. Many search protocols have been proposed for both structured and unstructured P2P networks, e.g., flooding, DHTs, and random walks. However, many of them don't consider the dynamics of P2P networks and the effects of topology on the searching processess. In this talk, we propose a referral-based approach for P2P networks, in which each peer is a software agent and the agents cooperate to search the whole P2P network through referrals. Our approach understands referrals as arising in and influencing dynamic P2P networks, where the agents act autonomously based on local knowledge. We study different strategies using which agents may search dynamic P2P networks. Further, we evaluate the proposed approach empirically for a community of AI scientists (partially derived from bibliographic data).

Related Readings:

Searching Social Networks Bin Yu, Munindar P. Singh In Proceedings of AAMAS, 2003 Making Gnutella-like P2P Systems Scalable Yatin Chawathe, Sylvia Ratnasamy, Lee Breslau, Nick Lanham, Scott Shenker In Proceedings of SIGCOMM, 2003 Peer-to-Peer Information Retrieval Using Self-Organizing Semantic Overlay Networks Chunqiang Tang, Zhichen Xu, Sandhya Dwarkadas In Proceedings of SIGCOMM, 2003