Gaussian traffic everywhere?

R. van de Meent, M.R.H. Mandjes, Aiko Pras

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

45 Citations (Scopus)
49 Downloads (Pure)

Abstract

It is often assumed that Internet traffic exhibits Gaussian characteristics, and this assumption has been validated in various studies of real Internet traffic. Less is known, however, about possible boundaries: at what timescales is traffic Gaussian and how much user aggregation is required for traffic to be Gaussian? The goal of this paper is to investigate these questions by analyzing hundreds of traffic traces, collected at four representative locations. To assess whether traffic is Gaussian, the paper starts with introducing an easy and fast procedure, based on earlier work of Kilpi and Norros. This procedure is used to investigate Gaussianity at timescales ranging from 5 msec to 5 sec. Our study shows that, if traffic is Gaussian at one timescale, it usually preserves this property at other timescales. The paper also investigates Gaussianity as function of the number of users. We conclude that, although it is impossible to give a hard number saying ‘above N users traffic is Gaussian’, it is fair to say that ‘only a few tens of users’ usually makes the aggregated traffic fairly Gaussian.
Original languageUndefined
Title of host publicationProceedings of the 2006 IEEE International Conference on Communications
Place of PublicationPiscataway, NJ, USA
PublisherIEEE Computer Society
Pages573-578
Number of pages6
ISBN (Print)1-4244-0355-3
DOIs
Publication statusPublished - 10 Jun 2006
EventIEEE International Conference on Communications, ICC 2006 - Istanbul, Turkey
Duration: 10 Jun 200614 Jun 2006

Publication series

Name
PublisherIEEE Computer Society
Number2

Conference

ConferenceIEEE International Conference on Communications, ICC 2006
Abbreviated titleICC
CountryTurkey
CityIstanbul
Period10/06/0614/06/06

Keywords

  • IR-65604
  • METIS-238008
  • EWI-2731

Cite this

van de Meent, R., Mandjes, M. R. H., & Pras, A. (2006). Gaussian traffic everywhere? In Proceedings of the 2006 IEEE International Conference on Communications (pp. 573-578). Piscataway, NJ, USA: IEEE Computer Society. https://doi.org/10.1109/ICC.2006.254856
van de Meent, R. ; Mandjes, M.R.H. ; Pras, Aiko. / Gaussian traffic everywhere?. Proceedings of the 2006 IEEE International Conference on Communications. Piscataway, NJ, USA : IEEE Computer Society, 2006. pp. 573-578
@inproceedings{313007f78ed24a39a98aea9247543b7b,
title = "Gaussian traffic everywhere?",
abstract = "It is often assumed that Internet traffic exhibits Gaussian characteristics, and this assumption has been validated in various studies of real Internet traffic. Less is known, however, about possible boundaries: at what timescales is traffic Gaussian and how much user aggregation is required for traffic to be Gaussian? The goal of this paper is to investigate these questions by analyzing hundreds of traffic traces, collected at four representative locations. To assess whether traffic is Gaussian, the paper starts with introducing an easy and fast procedure, based on earlier work of Kilpi and Norros. This procedure is used to investigate Gaussianity at timescales ranging from 5 msec to 5 sec. Our study shows that, if traffic is Gaussian at one timescale, it usually preserves this property at other timescales. The paper also investigates Gaussianity as function of the number of users. We conclude that, although it is impossible to give a hard number saying ‘above N users traffic is Gaussian’, it is fair to say that ‘only a few tens of users’ usually makes the aggregated traffic fairly Gaussian.",
keywords = "IR-65604, METIS-238008, EWI-2731",
author = "{van de Meent}, R. and M.R.H. Mandjes and Aiko Pras",
note = "10.1109/ICC.2006.254856",
year = "2006",
month = "6",
day = "10",
doi = "10.1109/ICC.2006.254856",
language = "Undefined",
isbn = "1-4244-0355-3",
publisher = "IEEE Computer Society",
number = "2",
pages = "573--578",
booktitle = "Proceedings of the 2006 IEEE International Conference on Communications",
address = "United States",

}

van de Meent, R, Mandjes, MRH & Pras, A 2006, Gaussian traffic everywhere? in Proceedings of the 2006 IEEE International Conference on Communications. IEEE Computer Society, Piscataway, NJ, USA, pp. 573-578, IEEE International Conference on Communications, ICC 2006, Istanbul, Turkey, 10/06/06. https://doi.org/10.1109/ICC.2006.254856

Gaussian traffic everywhere? / van de Meent, R.; Mandjes, M.R.H.; Pras, Aiko.

Proceedings of the 2006 IEEE International Conference on Communications. Piscataway, NJ, USA : IEEE Computer Society, 2006. p. 573-578.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Gaussian traffic everywhere?

AU - van de Meent, R.

AU - Mandjes, M.R.H.

AU - Pras, Aiko

N1 - 10.1109/ICC.2006.254856

PY - 2006/6/10

Y1 - 2006/6/10

N2 - It is often assumed that Internet traffic exhibits Gaussian characteristics, and this assumption has been validated in various studies of real Internet traffic. Less is known, however, about possible boundaries: at what timescales is traffic Gaussian and how much user aggregation is required for traffic to be Gaussian? The goal of this paper is to investigate these questions by analyzing hundreds of traffic traces, collected at four representative locations. To assess whether traffic is Gaussian, the paper starts with introducing an easy and fast procedure, based on earlier work of Kilpi and Norros. This procedure is used to investigate Gaussianity at timescales ranging from 5 msec to 5 sec. Our study shows that, if traffic is Gaussian at one timescale, it usually preserves this property at other timescales. The paper also investigates Gaussianity as function of the number of users. We conclude that, although it is impossible to give a hard number saying ‘above N users traffic is Gaussian’, it is fair to say that ‘only a few tens of users’ usually makes the aggregated traffic fairly Gaussian.

AB - It is often assumed that Internet traffic exhibits Gaussian characteristics, and this assumption has been validated in various studies of real Internet traffic. Less is known, however, about possible boundaries: at what timescales is traffic Gaussian and how much user aggregation is required for traffic to be Gaussian? The goal of this paper is to investigate these questions by analyzing hundreds of traffic traces, collected at four representative locations. To assess whether traffic is Gaussian, the paper starts with introducing an easy and fast procedure, based on earlier work of Kilpi and Norros. This procedure is used to investigate Gaussianity at timescales ranging from 5 msec to 5 sec. Our study shows that, if traffic is Gaussian at one timescale, it usually preserves this property at other timescales. The paper also investigates Gaussianity as function of the number of users. We conclude that, although it is impossible to give a hard number saying ‘above N users traffic is Gaussian’, it is fair to say that ‘only a few tens of users’ usually makes the aggregated traffic fairly Gaussian.

KW - IR-65604

KW - METIS-238008

KW - EWI-2731

U2 - 10.1109/ICC.2006.254856

DO - 10.1109/ICC.2006.254856

M3 - Conference contribution

SN - 1-4244-0355-3

SP - 573

EP - 578

BT - Proceedings of the 2006 IEEE International Conference on Communications

PB - IEEE Computer Society

CY - Piscataway, NJ, USA

ER -

van de Meent R, Mandjes MRH, Pras A. Gaussian traffic everywhere? In Proceedings of the 2006 IEEE International Conference on Communications. Piscataway, NJ, USA: IEEE Computer Society. 2006. p. 573-578 https://doi.org/10.1109/ICC.2006.254856