Home > Research > Publications & Outputs > Network Performance Implications of Variability...

Research

Electronic data

Links

Text available via DOI:

View graph of relations

Network Performance Implications of Variability in Data Traffic.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Network Performance Implications of Variability in Data Traffic. / Roadknight, Chris; Marshall, Ian W.; Bilchev, George.
In: BT Technology Journal, Vol. 18, No. 2, 04.2000, p. 151-158.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Roadknight, C, Marshall, IW & Bilchev, G 2000, 'Network Performance Implications of Variability in Data Traffic.', BT Technology Journal, vol. 18, no. 2, pp. 151-158. https://doi.org/10.1023/A:1026734014466

APA

Roadknight, C., Marshall, I. W., & Bilchev, G. (2000). Network Performance Implications of Variability in Data Traffic. BT Technology Journal, 18(2), 151-158. https://doi.org/10.1023/A:1026734014466

Vancouver

Roadknight C, Marshall IW, Bilchev G. Network Performance Implications of Variability in Data Traffic. BT Technology Journal. 2000 Apr;18(2):151-158. doi: 10.1023/A:1026734014466

Author

Roadknight, Chris ; Marshall, Ian W. ; Bilchev, George. / Network Performance Implications of Variability in Data Traffic. In: BT Technology Journal. 2000 ; Vol. 18, No. 2. pp. 151-158.

Bibtex

@article{eede487da2424609a20438a156e8aad0,
title = "Network Performance Implications of Variability in Data Traffic.",
abstract = "World Wide Web (WWW) traffic will dominate network traffic for the foreseeable future. Accurate predictions of network performance can only be achieved if network models reflect WWW traffic statistics. Through analysis of usage logs at a range of caches it is shown that WWW traffic is not a Poisson arrival process, and that it displays significant levels of self-similarity. It is also shown for the first time that the self-similar variability extends to demand for individual pages, and is far more pervasive than previously thought. These measurements are used as the basis for a cache-modelling tool-kit. Using this software the impact of the variability on predictive planning is illustrated. The model predicts that optimisations based on predictive algorithms (such as least recently used discard) are likely to reduce performance very quickly. This means that, far from improving the efficiency of the network, conventional approaches to network planning and engineering will tend to reduce efficiency and increase costs.",
author = "Chris Roadknight and Marshall, {Ian W.} and George Bilchev",
note = "The original publication is available at www.springerlink.com.",
year = "2000",
month = apr,
doi = "10.1023/A:1026734014466",
language = "English",
volume = "18",
pages = "151--158",
journal = "BT Technology Journal",
issn = "1573-1995",
publisher = "Kluwer Academic Publishers",
number = "2",

}

RIS

TY - JOUR

T1 - Network Performance Implications of Variability in Data Traffic.

AU - Roadknight, Chris

AU - Marshall, Ian W.

AU - Bilchev, George

N1 - The original publication is available at www.springerlink.com.

PY - 2000/4

Y1 - 2000/4

N2 - World Wide Web (WWW) traffic will dominate network traffic for the foreseeable future. Accurate predictions of network performance can only be achieved if network models reflect WWW traffic statistics. Through analysis of usage logs at a range of caches it is shown that WWW traffic is not a Poisson arrival process, and that it displays significant levels of self-similarity. It is also shown for the first time that the self-similar variability extends to demand for individual pages, and is far more pervasive than previously thought. These measurements are used as the basis for a cache-modelling tool-kit. Using this software the impact of the variability on predictive planning is illustrated. The model predicts that optimisations based on predictive algorithms (such as least recently used discard) are likely to reduce performance very quickly. This means that, far from improving the efficiency of the network, conventional approaches to network planning and engineering will tend to reduce efficiency and increase costs.

AB - World Wide Web (WWW) traffic will dominate network traffic for the foreseeable future. Accurate predictions of network performance can only be achieved if network models reflect WWW traffic statistics. Through analysis of usage logs at a range of caches it is shown that WWW traffic is not a Poisson arrival process, and that it displays significant levels of self-similarity. It is also shown for the first time that the self-similar variability extends to demand for individual pages, and is far more pervasive than previously thought. These measurements are used as the basis for a cache-modelling tool-kit. Using this software the impact of the variability on predictive planning is illustrated. The model predicts that optimisations based on predictive algorithms (such as least recently used discard) are likely to reduce performance very quickly. This means that, far from improving the efficiency of the network, conventional approaches to network planning and engineering will tend to reduce efficiency and increase costs.

U2 - 10.1023/A:1026734014466

DO - 10.1023/A:1026734014466

M3 - Journal article

VL - 18

SP - 151

EP - 158

JO - BT Technology Journal

JF - BT Technology Journal

SN - 1573-1995

IS - 2

ER -