SAIL (Statistically Accurate Internet Loss Measurements)

This is the homepage of SAIL, a rigorous statistical method to analyse the results of loss measurements. SAIL uses algorithms from Hidden Semi-Markov Models to estimate the parameters of the real loss process from measurement. It then computes the real loss rate and its variance. SAIL is light-weight and runs in linear time (on the number of samples). Below you can find the Matlab implementation of SAIL, together with the measurement data.

CODE

The complete MATLAB package can be downloaded here. The individual components are described below.

LIBRARY: this is the folder with all the subroutines for SAIL

runSim.m: Main program to generate simulated data with Gamma loss process and to run SAIL on simulated data
runExp.m: Main program to run SAIL on experiment data.
TESTSUIT: This folder contains tests that we have performed on the SAIL codes.
README: This file contains all instructions needed to run the code.

DATA

We have more than 50 GB of loss measurement data. Below are samples of the dataset in the text format that SAIL uses. If you want the complete set, please contact Hung Nguyen on hung(dot)nguyenATadelaide(dot)edu(dot)au.

PAPERS

"Rigorous statistical analysis of Internet loss measurements", H.X. Nguyen, M. Roughan, IEEE/ACM Transactions on Networking (TON) 21 (3), 734-745, 2013
"Statistically Accurate Network Measurements", Isaac Newton Institute Presention, 2010
“Statistically Accurate Internet Loss Measurement”, Hung X. Nguyen and Matthew Roughan, Technical Report, The University of Adelaide, Australia
“Statistically accurate Internet loss measurements”, H. X. Nguyen and M. Roughan, in Proc. of ACM Sigmetrics 2010 (short paper), New York, USA.
Presentation
“On the Correlation of Internet Packet Losses”, H. X. Nguyen and M. Roughan, in the Australasian Telecommunications Networks and Applications Conference (ATNAC), pp. 22–27, Adelaide, Australia, 7-10 December 2008.