Path-Aware multicast for efficient file distribution in peer-to-peer overlay networks

Authors:
Chun-Hsin Wu;Jia-Wei Li;Yueh-Ju Chen;Jum-Ping Lin
Affiliations:
Department of Computer Science and Information Engineering, National University of Kaohsiung, Taiwan;Department of Computer Science and Information Engineering, National University of Kaohsiung, Taiwan;Department of Computer Science and Information Engineering, National University of Kaohsiung, Taiwan;Department of Computer Science and Information Engineering, National University of Kaohsiung, Taiwan
Venue:
ICOIN'06 Proceedings of the 2006 international conference on Information Networking: advances in Data Communications and Wireless Networks
Year:
2006

Citing 4
Cited 0

A case for end system multicast (keynote address)

Proceedings of the 2000 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Resilient overlay networks

SOSP '01 Proceedings of the eighteenth ACM symposium on Operating systems principles
Distributing streaming media content using cooperative networking

NOSSDAV '02 Proceedings of the 12th international workshop on Network and operating systems support for digital audio and video
On Peer-to-Peer Media Streaming

ICDCS '02 Proceedings of the 22 nd International Conference on Distributed Computing Systems (ICDCS'02)

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper discusses how to transmit the same file or the same part of a large file from a source peer to a group of destination peers efficiently Considering the quality of overlay links and the restriction of Internet firewalls, we propose a hybrid approach of amplification and multicast, called amplicast, to improve the efficiency of file distribution By intelligent peer selection and the support of replicate peers, the proposed amplicast approach constructs a series of multicast sub-trees to alleviate the load of the source peer and allow a requesting peer to decide when and where to join a tree, based on its own benefits With link status cache and top-set sampling heuristic, the proposed PeerTop can significantly reduce the cost of exhausted network probing Experiments with real-world Internet traffic data demonstrate that our approach can distribute contents very efficiently.