Modeling and caching of peer-to-peer traffic

Osama Saleh, Mohamed Hefeeda

Research output: Chapter in Book/Report/Conference proceedingConference contribution

86 Citations (Scopus)

Abstract

Peer-to-peer (P2P) file sharing systems generate a major portion of the Internet traffic, and this portion is expected to increase in the future. We explore the potential of deploying proxy caches in different Autonomous Systems (ASes) with the goal of reducing the cost incurred by Internet service providers and alleviating the load on the Internet backbone. We conduct a measurement study to model the popularity of P2P objects in different ASes. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, regardless of the AS. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, less than 10% of the total traffic, a byte hit rate of up to 35% can be achieved by our algorithm, which is close to the byte hit rate achieved by an off-line optimal algorithm with complete knowledge of future requests. Our results also show that our algorithm achieves a byte hit rate that is at least 40% more, and at most triple, the byte hit rate of the common web caching algorithms. Furthermore, our algorithm is robust in face of aborted downloads, which is a common case in P2P systems.

Original languageEnglish
Title of host publicationProceedings - International Conference on Network Protocols, ICNP
Pages249-258
Number of pages10
DOIs
Publication statusPublished - 1 Dec 2006
Externally publishedYes
Event14th IEEE International Conference on Network Protocols, ICNP 2006 - Santa Barbara, CA, United States
Duration: 12 Nov 200615 Nov 2006

Other

Other14th IEEE International Conference on Network Protocols, ICNP 2006
CountryUnited States
CitySanta Barbara, CA
Period12/11/0615/11/06

Fingerprint

Proxy caches
Internet
Internet service providers
Costs

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Saleh, O., & Hefeeda, M. (2006). Modeling and caching of peer-to-peer traffic. In Proceedings - International Conference on Network Protocols, ICNP (pp. 249-258). [4110297] https://doi.org/10.1109/ICNP.2006.320218

Modeling and caching of peer-to-peer traffic. / Saleh, Osama; Hefeeda, Mohamed.

Proceedings - International Conference on Network Protocols, ICNP. 2006. p. 249-258 4110297.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Saleh, O & Hefeeda, M 2006, Modeling and caching of peer-to-peer traffic. in Proceedings - International Conference on Network Protocols, ICNP., 4110297, pp. 249-258, 14th IEEE International Conference on Network Protocols, ICNP 2006, Santa Barbara, CA, United States, 12/11/06. https://doi.org/10.1109/ICNP.2006.320218
Saleh O, Hefeeda M. Modeling and caching of peer-to-peer traffic. In Proceedings - International Conference on Network Protocols, ICNP. 2006. p. 249-258. 4110297 https://doi.org/10.1109/ICNP.2006.320218
Saleh, Osama ; Hefeeda, Mohamed. / Modeling and caching of peer-to-peer traffic. Proceedings - International Conference on Network Protocols, ICNP. 2006. pp. 249-258
@inproceedings{1e76757661314f409386d3b03e0a0318,
title = "Modeling and caching of peer-to-peer traffic",
abstract = "Peer-to-peer (P2P) file sharing systems generate a major portion of the Internet traffic, and this portion is expected to increase in the future. We explore the potential of deploying proxy caches in different Autonomous Systems (ASes) with the goal of reducing the cost incurred by Internet service providers and alleviating the load on the Internet backbone. We conduct a measurement study to model the popularity of P2P objects in different ASes. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, regardless of the AS. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, less than 10{\%} of the total traffic, a byte hit rate of up to 35{\%} can be achieved by our algorithm, which is close to the byte hit rate achieved by an off-line optimal algorithm with complete knowledge of future requests. Our results also show that our algorithm achieves a byte hit rate that is at least 40{\%} more, and at most triple, the byte hit rate of the common web caching algorithms. Furthermore, our algorithm is robust in face of aborted downloads, which is a common case in P2P systems.",
author = "Osama Saleh and Mohamed Hefeeda",
year = "2006",
month = "12",
day = "1",
doi = "10.1109/ICNP.2006.320218",
language = "English",
isbn = "1424405939",
pages = "249--258",
booktitle = "Proceedings - International Conference on Network Protocols, ICNP",

}

TY - GEN

T1 - Modeling and caching of peer-to-peer traffic

AU - Saleh, Osama

AU - Hefeeda, Mohamed

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Peer-to-peer (P2P) file sharing systems generate a major portion of the Internet traffic, and this portion is expected to increase in the future. We explore the potential of deploying proxy caches in different Autonomous Systems (ASes) with the goal of reducing the cost incurred by Internet service providers and alleviating the load on the Internet backbone. We conduct a measurement study to model the popularity of P2P objects in different ASes. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, regardless of the AS. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, less than 10% of the total traffic, a byte hit rate of up to 35% can be achieved by our algorithm, which is close to the byte hit rate achieved by an off-line optimal algorithm with complete knowledge of future requests. Our results also show that our algorithm achieves a byte hit rate that is at least 40% more, and at most triple, the byte hit rate of the common web caching algorithms. Furthermore, our algorithm is robust in face of aborted downloads, which is a common case in P2P systems.

AB - Peer-to-peer (P2P) file sharing systems generate a major portion of the Internet traffic, and this portion is expected to increase in the future. We explore the potential of deploying proxy caches in different Autonomous Systems (ASes) with the goal of reducing the cost incurred by Internet service providers and alleviating the load on the Internet backbone. We conduct a measurement study to model the popularity of P2P objects in different ASes. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, regardless of the AS. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, less than 10% of the total traffic, a byte hit rate of up to 35% can be achieved by our algorithm, which is close to the byte hit rate achieved by an off-line optimal algorithm with complete knowledge of future requests. Our results also show that our algorithm achieves a byte hit rate that is at least 40% more, and at most triple, the byte hit rate of the common web caching algorithms. Furthermore, our algorithm is robust in face of aborted downloads, which is a common case in P2P systems.

UR - http://www.scopus.com/inward/record.url?scp=46149087784&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46149087784&partnerID=8YFLogxK

U2 - 10.1109/ICNP.2006.320218

DO - 10.1109/ICNP.2006.320218

M3 - Conference contribution

AN - SCOPUS:46149087784

SN - 1424405939

SN - 9781424405930

SP - 249

EP - 258

BT - Proceedings - International Conference on Network Protocols, ICNP

ER -