Arvind Krishnamurthy

University of Washington

The Internet has emerged over the past decade as a critical piece of telecommunications infrastructure, yet the measured availability of the Internet is at least two orders of magnitude less than the telephone network.  As there is no single silver bullet, we have been looking at how to attack the problem both architecturally and through immediately deployable diagnostic services.  For outages that occur over short time scales, we address availability problems at the network protocol layer, by developing provably more reliable algorithms for routing using techniques from the theoretical distributed systems community.  For long-term outages, we address the problems at the management layer, by using timely alerts and automated diagnosis to identify connectivity problems so that ISPs can address them.

4:15pm

B17 Upson Hall

Tuesday, March 29, 2011

Refreshments at 3:45pm in the Upson 4th Floor Atrium

 

Computer Science

Colloquium

Spring 2011

www.cs.cornell.edu/events/colloquium

Towards a More Resilient Internet