The Internet is made up of over 35,000 smaller networks, owned by different economic entities (e.g. AT&T, Google). The Border Gateway Protocol (BGP) establishes routes between these networks and can be regarded as the “glue” that holds today’s Internet together.
BGP was designed at a time when the Internet was meant to provide connectivity
between largely trusted and cooperative parties. However, times have changed and today’s BGP
is notoriously vulnerable to attacks. To remedy this, secure variants of BGP have been proposed to prevent the propagation of bogus routing information. Unfortunately, despite a decade of extensive work the problem of securing the Internet’s interdomain routing is far from solved and deployment of a secure routing protocol is not on the horizon.
It is now clear that the two biggest impediments on the path to secure Internet routing are:
1. Which secure protocol to deploy? There are still lingering disagreements
about which of the security-enhanced variants of BGP should be deployed.
2. How to create economic incentives for deployment? Even once an agreement about which secure BGP variant to deploy will be reached, how can we get the ball rolling on protocol deployment?
My proposed research aims to inform this discussion and will consist of three main components: (1) investigating the security guarantees of the major proposed secure BGP variants; (2) exploring the vulnerabilities of the routing system to new kinds of attacks; and (3) designing market mechanisms for large-scale deployment of a secure routing protocol.
To achieve these goals, I plan to combine theoretical analysis with extensive simulations on real-life data. AIongside its practical contributions, the proposed research will involve posing and tackling new and exciting theoretical questions, motivated by Internet routing (topics on the borderline of distributed computing and game theory, non-local influence in social neworks, and more).
Fields of science
Call for proposal
See other projects for this call