Self Similarity Presentations and Programs

Ashish Gupta


PPTs

On the Self-Similar Nature of Ethernet Traffic

Given on April 23rd 2003

 


 

Other Presentations:

Scaling Phenomenon and Data Network Traffic : Slides by Willinger

Self-Similar through High-Variability: Statistical Analysis of Ethernet LAN Traffic at the Source Level : Presented at SIGCOMM by Willinger,...

Network Traffic Characteristics

Where Mathematics meets the Internet : Willinger and Paxson

Self-Similar Network Traffic

Network Traffic Characteristic

Self Similarity by Gustavo Serena - another ppt

On Multimedia Networks: Self-Similar Traffic and Network Performance : Good introduction to self-similarity and its implications

Applications of Large Deviations to Performance Analysis with Long-Range Dependent Traffic


 

Programs

Selfis : Program for analyzing self-similar data and computing H using various methods (JAVA Based)

Pareto : Generate Pareto distribution and also compare with Exponential Distribution (Windows Platform)

Heavy-tail plot : A simple Perl script to plot a log-log probability distribution of input data (outputs a PNG file) and also estimates alpha value from the data. Based on the aest tool. 

Aggregate: Program to aggregate data to desired levels. Useful for Self-similarity analysis.

Autocorrelation : Program to compute autocorrelation for a given set of data


 

Other Useful Links

ANURAN - Automatic Non-Uniform RANdom number generators

Fractional Gaussian Noise Applet

This web site provides links and very brief summaries for some of the work that has been done in the areas of self-similarity and long range dependence in computer networks.

Articles

Fractals, Heavy-Tails, and the Internet : Intro

Self Similarity : Graphical Intro

Digressive but interesting discussion of self similarity

The Self-Similar Traffic Modeling in the Internet

Totally Unscientific Study of Self-Similarity

Heavy-tailed Distributions : a brief intro

Zipf, Power-laws, and Pareto - a ranking tutorial

Long-range dependence and heavy-tail modeling for teletraffic data