Title :
Optimal Superpeer Selection Based on Load Balance for P2P File-sharing System
Author :
Gao, Lu ; Min Peng
Author_Institution :
State Key Lab. of Inf. Eng. in Surveying, Mapping & Remote Sensing, Wuhan Univ., Wuhan, China
Abstract :
The peer-to-peer (P2P) systems have many applications due to their virtues: high scalability and low cost. The systems using super-peers (SPs), which deal with all queries, instead of ordinary-peers (OPs), run more efficiently. So how to choose SPs from all peers is a key, but also a difficult problem because of the systems´ heterogeneity and dynamic. In this paper a novel mechanism for the selection of SP is proposed. The selection strategy is based on the dynamic distribution of peers and files in all clusters, and the capacity of the peer is taken into consideration. Through the simulation, the result shows the selection efficiency is improved and the load unbalance among SPs is relieved.
Keywords :
peer-to-peer computing; resource allocation; P2P file-sharing system; dynamic distribution; load balance; optimal superpeer selection; peer-to-peer systems; Artificial intelligence; Computer networks; Cost function; Network topology; Peer to peer computing; Position measurement; Power measurement; Remote sensing; Routing; Scalability; file distribution; load balance; peer-to-peer; super-peer selection;
Conference_Titel :
Artificial Intelligence, 2009. JCAI '09. International Joint Conference on
Conference_Location :
Hainan Island
Print_ISBN :
978-0-7695-3615-6
DOI :
10.1109/JCAI.2009.165
