DocumentCode :
3022390
Title :
A Category Overlay Infrastructure for Peer-to-Peer Content Search
Author :
Liu, Xin ; Wang, Jun ; Vuong, Son T. ; Son T Vuong
Author_Institution :
Dept. of Comput. Sci., British Columbia Univ., Vancouver, BC, Canada
fYear :
2005
fDate :
04-08 April 2005
Abstract :
For a Peer-to-Peer (P2P) system which contains huge amount of data, an efficient content search technique is definitely necessary. In this paper, we present a novel category overlay search infrastructure which can sit on top of existing unstructured P2P systems, such as Gnutella. The intuition behind our work is that by clustering the whole peer group into clusters and mapping the predefined category domain to clusters, multiple category overlays can be constructed and a search can be restricted within only some specific overlays, thus the search efficiency can be improved dramatically. Our preliminary simulation shows that this infrastructure can provide efficient search service, even in a highly dynamic network environment.
Keywords :
information retrieval; peer-to-peer computing; category overlay search infrastructure; content search technique; peer-to-peer system; Clustering algorithms; Computer science; Costs; Distributed processing; Floods; Keyword search; Maintenance; Network topology; Peer to peer computing; Usability;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Parallel and Distributed Processing Symposium, 2005. Proceedings. 19th IEEE International
Print_ISBN :
0-7695-2312-9
Type :
conf
DOI :
10.1109/IPDPS.2005.3
Filename :
1420103
Link To Document :
بازگشت