CSSE Seminar Series (CSSESS)

Quick links: Past seminars,  future seminars,  CSSESS Home

The transmission capacity of today's high-speed networks is often greater than the capacity of an end-system (such as a server or a remote client) to consume the incoming data. The mismatch between the network and the end-system, which can be exacerbated by high end-system workloads, will result in incoming packets being dropped in the end-system at different points in the packet receiving process. In particular, a packet may be dropped in the network adapter(NIC), in the kernel ring buffer, and (for rate based protocols) in the socket buffer. To provide reliable data transfers, these losses require retransmissions, and high loss rates will result in longer download times. We focus on UDP based transport protocols, and address the question of how to accurately estimate the rate at which the end-system can consume data which minimizes the overall transfer time of a file. We propose a queueing network model of the packet receiving process in the end-system. It consists of a model of the NIC, a model of the kernel ring buffer and the protocol processing, and a model of the socket buffer from which the application process reads the data. We show that using these simple and approximate queueing models, we can accurately predict the effective end-system bottleneck rate that minimizes the file transfer time. We compare our protocol with PA-UDP, an end-system aware rate based transport protocol, and show that our approach performs better, particularly when the packet losses in the NIC and/or the kernel ring buffer are high. We also compare our approach to TCP, which invokes its congestion control algorithm when there are losses in the NIC and in the ring buffer. As the end-to-end delay increases, TCP experiences significant performance degradation compared to our reliable end-system aware rate based protocol.

Dipak Ghosal received his B.Tech. degree in electrical engineering from the Indian Institute of Technology, Kanpur, in 1983; his M.S. degree in computer science from the Indian Institute of Science, Bangalore, in 1985; and his Ph.D. degree in computer science from the University of Louisiana, Lafayette, in 1988. From 1990 to 1996, he was a member of the technical staff at Bellcore (now Telcordia), Red Bank, New Jersey. He is currently a Professor in the Department of Computer Science at the University of California, Davis. His research interests are in the areas of high-speed networks, IP telephony, wireless networks, and performance evaluation of computer and communication systems.

Quick links: Past seminars,  future seminars,  CSSESS Home