×

zbMATH — the first resource for mathematics

Empirical testing of the infinite source Poisson data traffic model. (English) Zbl 1048.62080
Summary: The infinite source Poisson model is a fluid queue approximation of network data transmission that assumes that sources begin constant rate transmissions of data at Poisson time points for random lengths of time. This model has been a popular one as analysts attempt to provide explanations for observed features in telecommunication data such as self-similarity, long range dependence and heavy tails.
We survey some features of this model in cases where transmission length distributions have (a) tails so heavy that means are infinite, (b) heavy tails with finite mean and infinite variance and (c) finite variance. We survey the self-similarity properties of various descriptor processes in this model and then present analyses of four data sets which show that certain features of the model are consistent with the data while others are contradicted.
The data sets are: 1) the Boston University 1995 study of web sessions, 2) the UC Berkeley home IP HTTP data collected in November 1996, 3) traces collected in end of 1997 at a Customer Service Switch in Munich, and 4) detailed data from a corporate Ericsson WWW server from October 1998.

MSC:
62M02 Markov processes: hypothesis testing
90B20 Traffic problems in operations research
Software:
longmemo
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Abry P., Self-similar network analysis and performance evolution pp 30– (2000)
[2] Abry P., IEEE Trans. Inform. Theory 44 pp 2– (1998) · Zbl 0905.94006 · doi:10.1109/18.650984
[3] Adler R.J., I.M.S. Lecture Notes-Monogr. Series pp 12– (1990)
[4] Adler R.J., A practical guide to heavy tails: statistical techniques and applications (1998) · Zbl 0901.00010
[5] Andersen A.T., Perform. Eval. 41 pp 67– (2000) · Zbl 1051.68513 · doi:10.1016/S0166-5316(00)00006-7
[6] Asmussen S., Applied Probability and Queues (1987)
[7] Barford P., World Wide Web, Special Issue on Characterization and Perform. Eval. 2 pp 15– (1999) · doi:10.1023/A:1019236319752
[8] Beirlant J., J. Am. Stat. Assoc. 91 pp 1659– (1996)
[9] Beran J., Statistics for long-memory processes (1994) · Zbl 0869.60045
[10] Billingsley P., 2nd Ed., in: Convergence of probability measures (1999) · Zbl 0944.60003 · doi:10.1002/9780470316962
[11] Cambanis S., IEEE Trans. Inform. Theory 41 pp 628– (1995) · Zbl 0820.60022 · doi:10.1109/18.382010
[12] Cohen J., J. Appl. Math. Stochast. Anal. 1 pp 255– (1998) · Zbl 0919.62104 · doi:10.1155/S1048953398000227
[13] Constantine A.G., J. R. Stat. Soc. B 56 pp 97– (1994)
[14] Crovella M., A practical guide to heavy tails: statistical techniques and applications (1998)
[15] Daubechies I., Ten lectures on wavelets (1992) · Zbl 0776.42018 · doi:10.1137/1.9781611970104
[16] Davis R., Ann. Appl. Probab. 6 pp 1191– (1996) · Zbl 0879.60053 · doi:10.1214/aoap/1035463328
[17] Durrett R., J. Stochast. Proc. Appl. 5 pp 213– (1977) · Zbl 0368.60028 · doi:10.1016/0304-4149(77)90031-X
[18] Embrechts P., Modelling extremal events for insurance and finance (1997) · Zbl 0873.62116 · doi:10.1007/978-3-642-33483-2
[19] Erramilli A., IEEE/ACM Trans. Network Comput. 4 pp 209– (1996) · Zbl 01936284 · doi:10.1109/90.491008
[20] Flandrin P., IEEE Trans. Inform. Theory 38 pp 910– (1992) · Zbl 0743.60078 · doi:10.1109/18.119751
[21] Feigin P., Extremes 1 pp 391– (1999) · Zbl 0933.62087 · doi:10.1023/A:1009946321478
[22] Gilbert A.C., IEEE Trans. Inform. Theory 45 (3) pp 971– (1999) · Zbl 0945.94001 · doi:10.1109/18.761336
[23] in press.Gogl, H. Data Acquisition and Data Analysis in ATM based Networks. Report 1998, Technical University of Munich, available athttp://wwwjessen.informatik.tu-muenchen.de/forschung/leistungsanalyse/ATM_Project/atm_index.html.,
[24] in press.Gogl, H. Traffic statistics and application profiles in ATM networks. Report 1998, Technical University of Munich, available athttp://wwwjessen.informatik.tu-muenchen.de/forschung/leistungsanalyse/ATM_Project/atm_index.html.,
[25] Gladyshev E.G., Theory Probab. Appl. 6 pp 52– (1961) · Zbl 0107.12601 · doi:10.1137/1106004
[26] Guérin C.A., J. Fourier Anal. Appl. 4 (6) pp 403– (2000) · Zbl 0958.60034 · doi:10.1007/BF02510146
[27] Hall P., Biometrika 80 pp 246– (1993) · Zbl 0769.62062 · doi:10.1093/biomet/80.1.246
[28] Hall P., J. Timer Ser. Anal. 15 (6) pp 587– (1994) · Zbl 0815.62060 · doi:10.1111/j.1467-9892.1994.tb00214.x
[29] Michael Harrison J., Brownian Motion and Stochastic Flow Systems (1985) · Zbl 0659.60112
[30] Heath D., Math. Oper. Res. 23 pp 145– (1998) · Zbl 0981.60092 · doi:10.1287/moor.23.1.145
[31] Heath D., Ann. Appl. Probab. 9 pp 352– (1999) · Zbl 1059.60505 · doi:10.1214/aoap/1029962746
[32] Hill B.M., Ann. Stat. 3 pp 1163– (1975) · Zbl 0323.62033 · doi:10.1214/aos/1176343247
[33] Hunt F., SIAM J. Appl. Math. 50 pp 307– (1990) · Zbl 0692.58018 · doi:10.1137/0150018
[34] Ibragimov L.A., Gaussian random processes (1978) · Zbl 0392.60037 · doi:10.1007/978-1-4612-6275-6
[35] Istas J., Ann. Inst. Henri Poincaré 28 pp 537– (1992)
[36] Istas J., Ann. Inst. Henri Poincaré 33 pp 407– (1997) · Zbl 0882.60032 · doi:10.1016/S0246-0203(97)80099-4
[37] Istas J., C. R. Acad. Sci., Tome 316 pp 495– (1993)
[38] Jelenković P., Adv. Appl. Probab. 31 pp 394– (1999) · Zbl 0952.60098 · doi:10.1239/aap/1029955141
[39] in press.Kaj, I. Convergence of scaled renewal processes to fractional Brownian motion. Technical report, available athttp://www.math.uu.se/ ikaj/papers.html, 1999.,
[40] Kent J.T., J. R. Stat. Soc. B 59 (3) pp 679– (1997)
[41] Klüppelberg C., Bernoulli 1 pp 125– (1995) · Zbl 0842.60030 · doi:10.2307/3318683
[42] Konstantopoulos T., Queueing Systems Theory Appl. 28 pp 215– (1998) · Zbl 0908.90131 · doi:10.1023/A:1019190821105
[43] Kratz M., Stochast. Models 12 pp 699– (1996) · Zbl 0887.62025 · doi:10.1080/15326349608807407
[44] Kurtz T., Stochastic Networks: Theory and Applications (1996)
[45] Lamperti J.W., Trans. Am. Math. Soc. pp 62– (1962) · doi:10.1090/S0002-9947-1962-0138128-7
[46] Leland W.E., IEEE/ACM Trans. Networking 2 pp 1– (1994) · Zbl 01936506 · doi:10.1109/90.282603
[47] Maulik K., J. Appl. Probab. 4 pp 671– (2002) · doi:10.1017/S0021900200021975
[48] in press.Maulik, K.; Resnick. S. Small and large time scale analysis of a network traffic model. Technical report, available atwww.orie.cornell.edu/trlist/trlist.html, 2001.,
[49] in press.Miokosch, T.; Resnick, S.; Rootzén, H.; Stegeman. A. Is network traffic approximated by stable Lévy motion or fractional Brownian motion? Preprint 1999, available as #1999:32 athttp://www.md.chalmers.se/Stat/Research/Preprints.,
[50] Nolan J.P., Maximum likelihood estimation and diagnostics for stable processes. Levy processes (2001) · Zbl 0971.62008
[51] Prabhu N.U., Stochastic Storage Processes: Queues, Insurance risk, Dams and Data Communication (1998) · Zbl 0888.60073 · doi:10.1007/978-1-4612-1742-8
[52] Reiss R.D., Statistical analysis of extreme values (1997) · Zbl 0880.62002 · doi:10.1007/978-3-0348-6336-0
[53] Resnick S.I., Extreme Values, Regular Variation and Point Processes (1987) · Zbl 0633.60001 · doi:10.1007/978-0-387-75953-1
[54] Resnick S.I., Ann. Stat. 25 pp 1805– (1997) · Zbl 0942.62097 · doi:10.1214/aos/1069362376
[55] Resnick S., Oper. Res. 45 pp 235– (1997) · Zbl 0890.90066 · doi:10.1287/opre.45.2.235
[56] Resnick S., Ann. Appl. Probab. 9 (3) pp 797– (1999) · Zbl 0959.62076 · doi:10.1214/aoap/1029962814
[57] Resnick C., Adv. Appl. Probab. 29 pp 271– (1997) · Zbl 0873.60021 · doi:10.2307/1427870
[58] Resnick S.I., Ann. Appl. Probab. 10 pp 753– (2000) · Zbl 1083.60521 · doi:10.1214/aoap/1019487509
[59] Resnick S., Extremes 3 (4) pp 145– (2000) · Zbl 0971.62049 · doi:10.1023/A:1009996916066
[60] Resnick S., J. Appl. Probab. 37 (2) pp 575– (2000) · Zbl 1028.90008 · doi:10.1239/jap/1014842560
[61] Riedi R.H., Long range dependence: theory and applications (2001)
[62] Riedi R.H., Self-similar Network Traffic and Performance Evaluation (2000)
[63] Samorodnitsky G., Stable non-Gaussian random processes (1994) · Zbl 0925.60027
[64] in press.Tajvidi N. Confidence Intervals and Accuracy Estimation for Heavy tailed Generalized Pareto Distribution. Part of PhD thesis, 1996, Chalmers, available athttp://www.mai.liu.se/ nataj/,
[65] Taqqu M., Comp. Commun. Rev. 27 (1997) · doi:10.1145/263876.263879
[66] Tewfik A.H., IEEE Trans. Inform. Theory 38 pp 904– (1992) · Zbl 0743.60079 · doi:10.1109/18.119750
[67] Veitch D., IEEE Trans. Inform. Theory. (1998)
[68] Whitt W., Math. Oper. Res. 5 pp 67– (1980) · Zbl 0428.60010 · doi:10.1287/moor.5.1.67
[69] Willinger W., Stat. Sci. 10 pp 67– (1995) · Zbl 1148.90310 · doi:10.1214/ss/1177010131
[70] Willinger W., Comput. Commun. Rev. 25 pp 100– (1995) · doi:10.1145/217391.217418
[71] Willinger W., IEEE/ACM Trans. Networking pp To appear– (1996)
[72] Wood A.T.A., J. Comput. Graphical Stat. 3 pp 409– (1994)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.