EAGER: Rewriting the Rules: How Programmable Switches Impact Internet Security

Every web page and video travels through switches, specialized computers that route data across the Internet. The latest programmable switches let engineers load custom logic and run it at full speed on Internet traffic. This raises a question: exactly what tasks can a commodity switch handle (when today's specialized hardware for jobs like Internet security costs a hundred times more)? This project will answer that question by addressing two major network challenges. The first is on access control (deciding which traffic to allow or block), and the second is resilient routing, keeping data flowing when parts of the network are blocked or controlled by an adversary. An "expressibility map" will be created, detailing on what these switches can and cannot do: which functions run at full speed, which run only at a cost in latency or memory, and which are impossible. The timing is pivotal, as next-generation chips and switches are being designed now (for example XSight's X2, selected by SpaceX for Starlink), and initial study shows what this emerging hardware cannot yet do and what it should. The expressibility map will be published openly; the resulting prototype systems, being dual-use, will be shared responsibly with researchers. This work lowers the cost of secure, resilient Internet infrastructure for network operators, especially where connectivity is unreliable, and builds research capacity at an EPSCoR land-grant institution (Oklahoma State University). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2626074 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Hrishikesh Bhattacharya | Institution: Oklahoma State University, STILLWATER, OK | Award Amount: $242,654 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2626074 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2626074.html