Quantitatively Measure Access Control Mechanisms across Different Operating Systems

Access control mechanisms (ACM) play a critical role in protecting operating systems from malicious attacks. A variety of ACMs have been proposed till date, including discretionary access control (DAC) and mandatory access control (MAC). However, it is often challenging to evaluate and compare the quality of protection (QoP) of ACMs, especially when they are deployed on different platforms. In this paper, we propose an approach to quantitatively measure and compare the quality of ACMs. We introduce the notion of vulnerability profiles to capture the weakness of ACMs in protecting against attacks, and the notion of vulnerability coefficients as a numeric and platform-independent measurement of the protection quality of ACMs. Based on these two notions, our approach applies the Grey System Theory and an independent vulnerability scoring system to infer complete vulnerability profiles and to calculate fair and objective vulnerability coefficients for ACMs. To our knowledge, our approach is the first attempt for quantitative measurement and comparison of ACMs across different operating systems. Based on our approach, we implement a prototype tool called ACVAL that leverages logic programming to characterize, evaluate, and compare access control mechanisms. We apply ACVAL to three mainstream ACMs, and results that ACVAL is effective in evaluating access control mechanisms across different operating systems, a feature particularly useful to administrators of heterogenous IT systems.