Research Area: Trustworthy Computing Platforms and Devices
In a broad sense, the notion of trustworthy computing refers to the ability to demonstrate that a computing platform, device and application have a set of desirable properties that are either verified, primarily by formal methods applied when the system is designed and implemented, or verifiable by efficient procedures employed as the system runs. It is this broad notion of trustworthy computing that we adopt in the Center’s research vision. However, for the purposes of illustrating specific research areas, our initial focus is on trustworthy security properties.
In today’s cyberspace, commodity systems, devices, and applications lack formal assurance of the secrecy and integrity of security-sensitive data. The size and complexity of these systems, devices, and applications suggest that we will not achieve the level of assurance necessary to guarantee the absence of security vulnerabilities within the next two to three decades, if ever. Even the best-engineered code contains bugs in proportion to its size, and available formal methods–while holding great promise for the long term–are plagued by scalability challenges. Yet the convenience and low cost of commodity systems, devices, and applications offer unmatched appeal for both users and developers, dictating that security-sensitive workloads will be run on these systems for many years to come. This situation highlights the need for techniques to achieve trustworthy computing spaces and subspaces on commodity systems, devices, and applications that offer strong guarantees of isolation and confinement and yet enable trustworthy communications among these spaces and subspaces. An initial research thrust will focus on developing basic trustworthy computing technologies for hosts and expanding them to critical network components; e.g., network routers, gateways, management services, embedded systems, and autonomous, remotely controlled devices and vehicles.