Dina Katabi - 5/24/04
CSAIL, MIT

Protecting Provider Networks Against Unexpected Failures and Traffic Spikes

Adaptive multi-path routing is a powerful traffic engineering tool; it allows an ISP to run its network at a lower maximum utilization while supporting the same traffic demands. Recent research has proposed a few off-line routing optimizers such as the OSPF weight optimizer and the multi-commodity flow optimizer for MPLS networks. Off-line routing optimizers take as input the traffic matrix, which is the long-term average estimate of traffic demands in the domain. As a result, offline routing optimizers are limited by errors in traffic matrix estimates, do not accommodate unpredictable real-time traffic spikes, and do not cope well with unexpected link failures.

In this talk, I will propose TeXCP, an on-line intra-domain routing optimizer. TeXCP reacts to traffic demands between an ingress-egress pair in real-time. In contrast to some of the prior work, TeXCP does not require any global node coordination; rather, it builds on ideas from the eXplicit Control Protocol (XCP) to ensure stability in spite of feedback delays and traffic variability. It is also easy to implement and deploy in today's Internet.

Bio
Dina Katabi is an Assistant Professor in the Department of Electrical Engineering and Computer Science and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT. She received her PhD and MS from MIT in 2003 and 1999, and her Bachelor of Science from Damascus University in 1995. Her doctoral dissertation won a Sprowls award and an ACM Honorable Mention award. She has won the best student paper award in SIGCOMM 2000. She is an associate member of the Sigma XI Scientific Research Society, and a co-chair of the SIGCOMM workshop on Practice and Theory of Incentives in Networked Systems (PINS).