Comparative analysis of Ethernet traffic statistical characteristics evaluation methods

User Rating:

/ 2

Details: Category: IT technologies; Last Updated on Thursday, 16 May 2013 10:17; Published on Thursday, 15 November 2012 15:19; Hits: 6205

SocButtons v1.4

Authors:

A.V. Korablev, State Higher Educational Institution “National Mining University”, Assistant Lecturer of the Department of Electronics and Computing Techniques, Dnіpropetrovsk, Ukraine

A.V. Serov, State Higher Educational Institution “National Mining University”, Assistant Lecturer of the Department of Electronics and Computing Techniques, Dnіpropetrovsk, Ukraine

Abstract:

Purpose. To find out Ethernet self-similar traffic frame size statistical distribution and give some recommendations concerning use of different methods of traffic probability characteristics estimation.

Methodology. An authentic research methodology has been presented. Aggregation procedure of initial data unification according to time scale constant step was put in its base. Traffic distribution further analysis and its characteristics estimation with different time interval lengths was carried out. Three ways of frame size distribution shape parameter evaluation were tested. Four approaches were applied to Hurst parameter estimation.

Findings. Tail index and Hurst parameter estimation results are presented.

Originality consists in self-similar traffic characteristics estimation methods analysis that allowed choosing the most effective method and giving recommendations concerning its usage.

Practical value. Recommendations on definite estimation method choice are given.

References:

1. Leland, W., Taqqu, M., Willinger, W. and Wilson, D. (1994), “On the self-similar nature of Ethernet traffic (extended version)”, IEEE/ACM Transactions on networking, vol.2, no.1. pp. 1–15.

2. “The Internet Traffic Archive”, available at: http://ita.ee.lbl.gov/html/contrib/BC.html

3. Adler, R.A., Feldman, R. and Taqqu, M. (1998), Practical guide to heavy tails: statistical techniques and applications, Birkhauser, Boston.

4. Crovella, M. and Taqqu, M. (1999), “Estimating the heavy-tail index from scaling properties”, Methodology and computing in applied probability, vol.1, no.1, pp. 55–79.

5. Grimm, C. and Schluchtermann, G. (2008), IP Traffic Theory and Performance, Springer-Verlag Berlin and Heidelberg, Berlin.

6. Востров Г.Н. Сегментация и анализ временных рядов на основе стохастической фрактальной модели / Г.Н. Востров, М.В. Полякова, В.В. Любченко // Тр. Одес. политехн. ун-та. – Одесса, 2001. –Вып. 1. – С. 109–114. – Библиогр.: с.114.

Vostrov, G.N., Polyakova, M.V. and Lyubchenko, V.V. (2001), “Segmentation and analysis of time series based on stochastic fractal model”, Proc. Odes. polytekhn. un-ta, Odessa, Issue 1. pp. 109–114.

7. Гуда А.Н. Модели оценки параметров телекоммуникационного трафика в автоматизированных информационно-управляющих системах / А.Н. Гуда, М.А. Бутакова, Н.А. Москат // Вопросы современной науки и практики. Ун-т им. В.И. Вернадского. – 2010. – №4–6(29). – С. 71–87.– Библиогр.: с.87.

Guda, A.N., Butakova, M.A. and Moskat, N.A. (2010), “Telecommunication traffic parameters estimation models in automated information-management systems”, Voprosy sovremennoy nauki i praktiki, Published by V.I. Vernadsky University, no.4–6(29), pp. 71–87.

Files:

2012_02_korabl


2013-04-26 365.22 KB 1608