Title :
Cluster: An Effective Solution to the Problem of Heavy-Tailed Distribution in P2P Networks
Author :
Wei, Guiyi ; Gu, Ye ; Ge, Yujia
Author_Institution :
Coll. of Comput. & Inf. Eng., Zhejiang Gongshang Univ., Hangzhou, China
fDate :
June 30 2009-July 2 2009
Abstract :
Unstructured peer-to-peer (P2P) network is a prevalent model in current P2P networks. In such systems, due to the existence of free-riders, some users cannot finally complete their download tasks for the scarcity of certain resources. Many unstructured P2P models divide download files into chunks. This paper introduces a working model called cluster to solve the heavy-tailed problem in file downloads. It analyses the distribution of file chunks and increases the replication speed of the rarest chunks transmitted among the group. Our simulation experiments show the proposed model can transmit the rarest chunk efficiently and improve the overall download completion rate.
Keywords :
peer-to-peer computing; P2P networks; file chunks distribution; heavy-tailed distribution; unstructured peer-to-peer network; Bandwidth; Clustering algorithms; Collaborative work; Computer networks; Distributed computing; Educational institutions; Heuristic algorithms; Peer to peer computing; Robustness; Scalability; P2P network; cluster; long-tail;
Conference_Titel :
New Trends in Information and Service Science, 2009. NISS '09. International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-0-7695-3687-3
DOI :
10.1109/NISS.2009.218
