DocumentCode :
3064801
Title :
Parallel Downloading Algorithm for Large-volume File Distribution
Author :
Chen, Haitao ; Gong, Zhenghu ; Huang, Zunguo
Author_Institution :
National University of Defense Technology
fYear :
2005
fDate :
05-08 Dec. 2005
Firstpage :
745
Lastpage :
749
Abstract :
BitTorrent is a popular, open-source content distribution system that is conducive for distribution of large-volume files. In this paper, we first model the parallel downloading problem as directed graph and discuss possible improving directions. Then a smart connection management algorithm is proposed, which introduces the notion of content group and sub-tracker, and decentralizes the decision process performed at the tracker for transfer coordination. Simulations show that by increasing content availability to the clients from their immediate neighbors, it can significantly improve the efficiency of distribution.
Keywords :
Availability; Bandwidth; Computer science; Content management; Environmental management; Internet; Open source software; Peer to peer computing; Research and development; Throughput;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Parallel and Distributed Computing, Applications and Technologies, 2005. PDCAT 2005. Sixth International Conference on
Print_ISBN :
0-7695-2405-2
Type :
conf
DOI :
10.1109/PDCAT.2005.184
Filename :
1579021
Link To Document :
بازگشت