Network Systems Group Princeton University   Publication Info

Title

Understanding TCP Vegas: A Duality Model

Authors

Steven Low
Larry Peterson
Limin Wang

Abstract

We view congestion control as a distributed primal-dual algorithm carried out by sources and links over a network to solve a global optimization problem. We describe a multi-link multi-source model of the TCP Vegas congestion control mechanism. The model provides a fundamental understanding of delay, fairness and loss properties of TCP Vegas. It implies that Vegas stabilizes around a weighted proportionally fair allocation of network capacity when there is sufficient buffering in the network. It clarifies the mechanism through which persistent congestion may arise and its consequences, and suggests how we might use REM active queue management to prevent it. We present simulation results that validate our conclusions.

Published
As

Journal of the ACM,
49(2):207-235, March 2002.
A preliminary version of this paper appeared at SIGMETRICS 2001.

Download

PostScript
PDF
BibTeX