Traffic modeling and proportional partial caching for peer-to-peer systems

Mohamed Hefeeda, Osama Saleh

Research output: Contribution to journalArticle

180 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 an eight-month measurement study to analyze the P2P traffic characteristics that are relevant to caching, such as object popularity, popularity dynamics, and object size. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, and that several workloads exist in P2P traffic. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and proportional partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, 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
Pages (from-to)1447-1460
Number of pages14
JournalIEEE/ACM Transactions on Networking
Volume16
Issue number6
DOIs
Publication statusPublished - 12 Aug 2008
Externally publishedYes

Fingerprint

Proxy caches
Internet
Internet service providers
Costs

Keywords

  • Internet measurement
  • Network protocols
  • Peer-to-peer systems
  • Traffic analysis
  • Traffic modeling

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Computer Science Applications
  • Computer Networks and Communications

Cite this

Traffic modeling and proportional partial caching for peer-to-peer systems. / Hefeeda, Mohamed; Saleh, Osama.

In: IEEE/ACM Transactions on Networking, Vol. 16, No. 6, 12.08.2008, p. 1447-1460.

Research output: Contribution to journalArticle

Hefeeda, Mohamed ; Saleh, Osama. / Traffic modeling and proportional partial caching for peer-to-peer systems. In: IEEE/ACM Transactions on Networking. 2008 ; Vol. 16, No. 6. pp. 1447-1460.
@article{a760794a39ae45269adc4ea718f290ba,
title = "Traffic modeling and proportional partial caching for peer-to-peer systems",
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 an eight-month measurement study to analyze the P2P traffic characteristics that are relevant to caching, such as object popularity, popularity dynamics, and object size. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, and that several workloads exist in P2P traffic. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and proportional partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, 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.",
keywords = "Internet measurement, Network protocols, Peer-to-peer systems, Traffic analysis, Traffic modeling",
author = "Mohamed Hefeeda and Osama Saleh",
year = "2008",
month = "8",
day = "12",
doi = "10.1109/TNET.2008.918081",
language = "English",
volume = "16",
pages = "1447--1460",
journal = "IEEE/ACM Transactions on Networking",
issn = "1063-6692",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",

}

TY - JOUR

T1 - Traffic modeling and proportional partial caching for peer-to-peer systems

AU - Hefeeda, Mohamed

AU - Saleh, Osama

PY - 2008/8/12

Y1 - 2008/8/12

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 an eight-month measurement study to analyze the P2P traffic characteristics that are relevant to caching, such as object popularity, popularity dynamics, and object size. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, and that several workloads exist in P2P traffic. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and proportional partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, 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 an eight-month measurement study to analyze the P2P traffic characteristics that are relevant to caching, such as object popularity, popularity dynamics, and object size. Our study shows that the popularity of P2P objects can be modeled by a Mandelbrot-Zipf distribution, and that several workloads exist in P2P traffic. Guided by our findings, we develop a novel caching algorithm for P2P traffic that is based on object segmentation, and proportional partial admission and eviction of objects. Our trace-based simulations show that with a relatively small cache size, 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.

KW - Internet measurement

KW - Network protocols

KW - Peer-to-peer systems

KW - Traffic analysis

KW - Traffic modeling

UR - http://www.scopus.com/inward/record.url?scp=57949085517&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57949085517&partnerID=8YFLogxK

U2 - 10.1109/TNET.2008.918081

DO - 10.1109/TNET.2008.918081

M3 - Article

VL - 16

SP - 1447

EP - 1460

JO - IEEE/ACM Transactions on Networking

JF - IEEE/ACM Transactions on Networking

SN - 1063-6692

IS - 6

ER -