CSCI B649: Trusted Computing
Instructor: Raquel Hill
Office: 230 E Lindley Hall
Office hours: TBD
We are building increasingly more complex applications to carry out some of the most critical tasks in our society. We have long understood that it is essential to validate computers, programs, or pieces of hardware, and we have myriad acceptable methods for performing those tasks. However, as we begin composing larger and larger systems out of what may well be individually trusted components, we have introduced a level of vulnerability in the dependencies between those components. These components include the operating systems, loadable kernel modules, device drivers, application run-time environments, etc. Currently, we do not have a systematic way of understanding all of the dependencies and relationships that may exist within a large software system, and until we develop a holistic, systematic way of understanding these trust dependencies, we can expect to have vulnerabilities in our computing infrastructure.
In this class, we will explore current methods for characterizing, establishing and attesting trust of a system. We will cover a variety of topics including hardware-based trusted components, trusted identities and identity management, reputation systems, trust negotiation, etc. Our goals for the course are to:
· Understand the various issues that require the ability to prove that a computer system can be trusted. The focus will be on the design of trustworthy systems and processes.
· Investigate a trusted computing problem through a semester long research project.
Grading Policy and Workload
· Class Participation 50%
· Project 50%
The course reading material will be made available online. We will review at least two research papers in detail during each class. Students are required to submit a critical review of the selected papers by noon the day of the class. Each class will involve a combination of: presentation of material by the instructor, student presentation(s), and the discussion of critical issues raised in the paper(s). The class participation part of the grade will be judged on: quality of participation in class discussions, class presentation(s), and critical review of papers.
The course project will be done in small teams and each team will work on a separate research problem. Students will have the flexibility to either choose from the problems suggested by the instructor or formulate their own with the help of the instructor. Projects will involve additional project-specific background reading and their progress will be monitored through weekly meetings throughout the semester. Each project is expected to produce a publication-quality report at completion. The project grade will be judged on: findings of the project, project report, and presentation of the project.