On the Characteristics of Internet Traffic Variability: Spikes and Elephants

Tatsuya MORI; Ryoichi KAWAHARA; Shozo NAITO; Shigeki GOTO

On the Characteristics of Internet Traffic Variability: Spikes and Elephants

Tatsuya MORI, Ryoichi KAWAHARA, Shozo NAITO, Shigeki GOTO

Full Text Views

0

Share
Cite this

Summary :

Analysing and modeling of traffic play a vital role in designing and controlling of networks effectively. To construct a practical traffic model that can be used for various networks, it is necessary to characterize aggregated traffic and user traffic. This paper investigates these characteristics and their relationship. Our analyses are based on a huge number of packet traces from five different networks on the Internet. We found that: (1) marginal distributions of aggregated traffic fluctuations follow positively skewed (non-Gaussian) distributions, which leads to the existence of "spikes", where spikes correspond to an extremely large value of momentary throughput, (2) the amount of user traffic in a unit of time has a wide range of variability, and (3) flows within spikes are more likely to be "elephant flows", where an elephant flow is an IP flow with a high volume of traffic. These findings are useful in constructing a practical and realistic Internet traffic model.

Publication: IEICE TRANSACTIONS on Information Vol.E87-D No.12 pp.2644-2653

Publication Date: 2004/12/01

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Section on New Technologies and their Applications of the Internet)

Category: Traffic Measurement and Analysis

Cite this

Copy

Tatsuya MORI, Ryoichi KAWAHARA, Shozo NAITO, Shigeki GOTO, "On the Characteristics of Internet Traffic Variability: Spikes and Elephants" in IEICE TRANSACTIONS on Information, vol. E87-D, no. 12, pp. 2644-2653, December 2004, doi: .
Abstract: Analysing and modeling of traffic play a vital role in designing and controlling of networks effectively. To construct a practical traffic model that can be used for various networks, it is necessary to characterize aggregated traffic and user traffic. This paper investigates these characteristics and their relationship. Our analyses are based on a huge number of packet traces from five different networks on the Internet. We found that: (1) marginal distributions of aggregated traffic fluctuations follow positively skewed (non-Gaussian) distributions, which leads to the existence of "spikes", where spikes correspond to an extremely large value of momentary throughput, (2) the amount of user traffic in a unit of time has a wide range of variability, and (3) flows within spikes are more likely to be "elephant flows", where an elephant flow is an IP flow with a high volume of traffic. These findings are useful in constructing a practical and realistic Internet traffic model.
URL: https://globals.ieice.org/en_transactions/information/10.1587/e87-d_12_2644/_p

Copy

@ARTICLE{e87-d_12_2644,
author={Tatsuya MORI, Ryoichi KAWAHARA, Shozo NAITO, Shigeki GOTO, },
journal={IEICE TRANSACTIONS on Information},
title={On the Characteristics of Internet Traffic Variability: Spikes and Elephants},
year={2004},
volume={E87-D},
number={12},
pages={2644-2653},
abstract={Analysing and modeling of traffic play a vital role in designing and controlling of networks effectively. To construct a practical traffic model that can be used for various networks, it is necessary to characterize aggregated traffic and user traffic. This paper investigates these characteristics and their relationship. Our analyses are based on a huge number of packet traces from five different networks on the Internet. We found that: (1) marginal distributions of aggregated traffic fluctuations follow positively skewed (non-Gaussian) distributions, which leads to the existence of "spikes", where spikes correspond to an extremely large value of momentary throughput, (2) the amount of user traffic in a unit of time has a wide range of variability, and (3) flows within spikes are more likely to be "elephant flows", where an elephant flow is an IP flow with a high volume of traffic. These findings are useful in constructing a practical and realistic Internet traffic model.},
keywords={},
doi={},
ISSN={},
month={December},}

Copy

TY - JOUR
TI - On the Characteristics of Internet Traffic Variability: Spikes and Elephants
T2 - IEICE TRANSACTIONS on Information
SP - 2644
EP - 2653
AU - Tatsuya MORI
AU - Ryoichi KAWAHARA
AU - Shozo NAITO
AU - Shigeki GOTO
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E87-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2004
AB - Analysing and modeling of traffic play a vital role in designing and controlling of networks effectively. To construct a practical traffic model that can be used for various networks, it is necessary to characterize aggregated traffic and user traffic. This paper investigates these characteristics and their relationship. Our analyses are based on a huge number of packet traces from five different networks on the Internet. We found that: (1) marginal distributions of aggregated traffic fluctuations follow positively skewed (non-Gaussian) distributions, which leads to the existence of "spikes", where spikes correspond to an extremely large value of momentary throughput, (2) the amount of user traffic in a unit of time has a wide range of variability, and (3) flows within spikes are more likely to be "elephant flows", where an elephant flow is an IP flow with a high volume of traffic. These findings are useful in constructing a practical and realistic Internet traffic model.
ER -