TCP Vegas: New Techniques for Congestion Detection and Avoidance

L. Brakmo, S. O'Malley and L. Peterson
Proceedings of the SIGCOMM '94 Symposium, August 1994, pg. 24-35

Vegas is a new implementation of TCP that achieves betwen 40 and 70% better throughtput, with one-fifth to one half the losses, as compared to the implementation of TCP in the Reno distribution of BSD Unix. This paper motivates and describes the three key techniques employed by Vegas, and presents the results of a comprehensive experimental performance study --using both simulation and measurements on the Internet-- of the Vegas and Reno implementations of TCP.

Performance Problems in BSD4.4 TCP

L. Brakmo and L. Peterson
Computer Communication Review, Vol 25, No. 5, October 1995, pg. 69-86.
143KB postscript file available here.

This paper describes problems in the BSD 4.4-Lite version of TCP (some of which are also present in earlier versions, such as the Net2 implementation of TCP) and proposes fixes that result in a 21% increase in throughput under realistic conditions.

TCP Vegas: End to End Congestion Avoidance on a Global Internet

L. Brakmo and L. Peterson
IEEE Journal on Selected Areas in Communication, Vol 13, No. 8, October 1995, pg. 1465-1480.
214KB postscript file available here.

Extended journal version of the SIGCOMM '94 paper. Includes new techniques for congestion avoidance during slow-start, extended simulation and analysis as well as expanded sections dealing with fairness, stability and queue behavior.

Experiences with Network Simulation

L. Brakmo and L. Peterson
SIGMETRICS '96, to appear.
180KB postscript file available here.

Simulation is a critical tool in developing, testing, and evaluating network protocols and architectures. This paper describes x-Sim, a network simulator based on the x-kernel, that is able to fully simulate the topologies and traffic patterns of large scale networks. It also gives case studies to help illustrate the capabilities and usefulness of the simulator. Finally, we consider a set of basic simulation principles in the context of running network simulations, giving a set of concrete guidelines and, most importantly, providing examples to help quantify the value of these principles as well as the cost of ignoring them.

Back to Lawrence's home page