HASSAYOUN Sofiane1, ROS SANCHEZ David1
Collection des rapports de recherche & preprints de TELECOM Bretagne
Brest : Institut TELECOM/TELECOM Bretagne, 2008, (Collection des rapports de recherche & preprints de TELECOM Bretagne, RR-2008001-RSM, ISSN 1255-2275)
In recent years, a good deal of research work has been devoted to studying and improving the performance of the Transmission Control Protocol (TCP) over very high-speed links. On the one hand, the issue of how to adapt TCP's congestion control mechanisms to Gb/s speeds and beyond has given rise to several protocol proposals. On the other hand, starting from the seminal work by Appenzeller et al. (2004), quite a few papers have dealt with the problem of how to adequately dimension router buffers for links running at very high speeds. Several conflicting rules have thus been proposed in the literature to compute the "right" amount of buffering. In this paper we are interested in looking at packet loss synchronization, and the related issue of buffer sizing, when the bulk of the traffic is composed of a moderate number of flows using high-speed versions of TCP. In particular, we wanted to explore whether increased levels of synchronization would be an outcome of using more aggressive versions of TCP over high-bandwidth paths. By means of ns-2 simulations, we have evaluated several TCP versions and a wide range of buffer sizes, as well as three different drop synchronization metrics. Our preliminary findings suggest that high-speed versions of TCP do yield higher levels of synchronization. However, in spite of a strong drop synchronization, such TCP versions can achieve both high goodput and link utilization, as long as enough buffering is provided.
1 : RSM(TB) - Dépt. Réseaux, Sécurité et Multimédia (Institut Mines-Télécom-Télécom Bretagne-UEB)
TCP, Congestion control, Loss synchronization, Buffer sizing