Abstract
Purchase PDF (2165 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (2)
Purchase PDF (507 K)
doi:10.1016/j.peva.2007.12.003 
Copyright © 2007 Elsevier Ltd All rights reserved.
References and further reading may be available for this article. To view references and further reading you must purchase this article.
Received 26 October 2007;
Abstract
In this paper we propose a method, related to the theory of Network Calculus, for the analysis of aggregate network traffic by means of empirical rate-interval curves obtained from experimental data. The algorithm on which it is based differs from the commonly employed wavelet-based approach, although it retains some of its multiresolution features. We briefly introduce the theoretical aspects, analyze measurement accuracy and present results, obtained both by simulation and by the analysis of real traffic traces, which provide an assessment of the strengths and weaknesses of the proposed method.
Rate-interval curve analysis provides very robust and acceptably accurate estimates of the Hurst parameter value and, even in the presence of flow irregularities, results can be proved to be correct as far as scaling properties are concerned.
Further analyzes concerning peaks, bursts and similar localized phenomena that may have a significant impact on the performances of a network are allowed by considering maximal rate envelopes, showing the potential of this approach for monitoring applications.
Keywords: Traffic measurement; Flow rate; Uncertainty evaluation
Article Outline
- 1. Introduction
- 2. Traffic analysis by rate-interval curves
- 2.1. Aggregate traffic and scale invariance
- 2.2. Theoretical analysis
- 2.3. RIC estimation algorithm
- 2.4. Complexity analysis
- 3. RIC log-scale diagram and Hurst parameter estimation
- 4. Validation with fractional Brownian motion traffic models
- 5. Modelling, heavy tails and violation probability
- 6. Analysis of experimental traffic data
- 7. Traffic monitoring example
- 8. Conclusion
- References
- Vitae
