Title: A model-driven emulation approach to large-scale TCP performance evaluation

Authors: Miguel A. Erazo, Jason Liu

Addresses: School of Computer and Information Sciences, Florida International University, Miami, Florida 33199, USA. ' School of Computer and Information Sciences, Florida International University, Miami, Florida 33199, USA

Abstract: Small-scale experiments are insufficient for a comprehensive study of congestion control protocols. Similarly, results obtained from pure simulation platforms without exercising real protocols and applications lack realism. Motivated by these reasons, we developed scalable virtualised evaluation environment for TCP (SVEET), a TCP performance evaluation testbed where real implementations of TCP variants can be accurately evaluated under diverse network configurations and workloads from real applications in large-scale network settings. It is our purpose to provide the research community a standard environment through which quantitative assessment regarding TCP behaviour can be drawn from large-scale experiments. In order to accomplish our goal, we adopted a novel model-driven emulation approach combining real-time simulation, machine and time virtualisation techniques. We validate the testbed via extensive experiments to assess its potentials and limitations. Additionally, we performed case studies involving real web, streaming and peer-to-peer applications. Our results indicate that SVEET can accurately model the behaviour of the TCP variants and support large-scale network scenarios.

Keywords: TCP performance; transmission control protocol; network simulation; model-driven emulation; machine virtualisation; time dilation; performance evaluation; congestion control protocols; network testbeds; web applications; streaming; peer-to-peer; P2P applications.

DOI: 10.1504/IJCNDS.2010.033971

International Journal of Communication Networks and Distributed Systems, 2010 Vol.5 No.1/2, pp.130 - 150

Published online: 05 Jul 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article