USA - 2016
For innovations that improved the efficiency of the Border Gateway Protocol (BGP) in routing Internet traffic, for laying the groundwork for software-defined networks (SDNs), and for contributions in measuring and engineering IP networks.Press Release
USA - 2008
For contributions to network control and management systems.
USA - 2004
For models, algorithms, and deployed systems that assure stable and efficient Internet routing without global coordination.
The Border Gateway Protocol (BGP) is the glue that holds the disparate parts of the Internet together. However, the rapid commercialization of the Internet in the last decade has put a significant strain on the interdomain-routing system. Unlike traditional routing protocols, BGP allows the operators of each domain to design policies for selecting paths to reach the Internet's many destinations and deciding which other networks may use these paths.
BGP offers significant flexibility in selecting policies, but these policies have (at best) an indirect influence on where the traffic goes. Jennifer Rexford's work has addressed these problems through deep understanding and innovation across Computer Science theory, networking protocols, Internet economics, and ISPs' operational practices.
Dr. Rexford's work with Lixin Gao was the first to recognize that business relationships in the Internet have a profound influence on the behavior of the underlying protocols. This work identified simple policy guidelines that ensure the global routing system converges while remaining faithful to the economic incentives of each domain. Dr. Rexford's work with Nick Feamster provided an efficient algorithm for modeling the effects of BGP policy changes on the flow of traffic through a domain. The operators of each domain can use the algorithm to select routing policies that maximize user performance and balance network load. Hence Dr. Rexford's work identified how each domain should constrain its policies to meet the overarching goal of the stability of the Internet, as well as the concrete methods each domain can apply to satisfy its own engineering goals within those constraints.