Leland, Will E.
Taqq, Murad S.
Willinger, Walter
Wilson, Daniel V.
Abstract
| We demonstrate that Ethernet local area network (LAN) traffic is
statistically self-similar, that none of the commonly used traffic
models is able to capture this fractal behavior, that such behavior
has serious implications for the design, control, and analysis of
high-speed, cell-based (B-ISDN) networks, and that aggregating
streams of such traffic typically intensifies the selfsimilarity
("burstiness") instead of smoothing it. Intuitively, the critical
characteristic of this self-similar traffic is that there is no
natural length of a "burst": at every time scale ranging from a
few milliseconds to minutes and hours, similar-looking traffic
bursts are evident. Our conclusions are supported by a rigorous
statistical analysis of hundreds of millions of high quality Ethernet
traffic...
|
Keywords
self similar
bursty
ethernet
Notes
The original paper on self similar traffic
- Studied several high quality ethernet traces
- Observed self-similar or fractal like traffic
- Different from pure poisson, poisson related or fluid models
- Variable Bit Rate traffic has similar properties to LAN traffic
- Congestion produced by self similar traffic is different and more complex
than that produced by previous models
- External TCP traffic offered to the LAN had the same self similar properties
|
Related Papers
Bibtex
@inproceedings{ leland_ether93,
author = "Will E. Leland and Murad S. Taqq and Walter Willinger and Daniel V. Wilson",
title = "On the self-similar nature of {Ethernet} traffic",
booktitle = "ACM SIGCOMM",
address = "San Francisco, California",
editor = "Deepinder P. Sidhu",
url = "ftp://thumper.bellcore.com/pub/wel/sigcomm93.ps.Z",
pages = "183--193",
year = "1993"
}