Abstract

Fraser Research is a not-for-profit institute for clean slate networking research based in Princeton, NJ. It aims to conduct leading edge, long term, fundamental research in network communications and to broaden and enrich the education of graduate students.

Security is a primary concern. The Internet has brought vulnerability to consumers through their use of computers, and to nations through the automation of their infrastructures. Attacks are launched from around the world and there is pessimism among experts that these attacks can be blocked. Today, citizens across the United States are inundated with junk mail and many are being robbed by invisible hands. As networked computing becomes deeply embedded in national infrastructures and critical services, so the risk of national disaster becomes an inescapable concern.

Much of the present problem stems from insecure host software, or from computer users who do not have adequate training to operate their computers securely. This has led some people to blame the users rather than the network. However, a country cannot expect its citizens to deploy consumer products with military grade security. Nor can it expect consumer products to be manufactured with military standards. A more reasonable solution begins with a network which diminishes the risk to its users by itself being robust. Follow that with a service delivery method that can easily be deployed by ordinary citizens.

In this talk, we will present an overview of the Fraser Research network architecture that we worked on over the course of the summer and detail how a clean slate design can avoid the mistakes the Internet has made.