Quantitative Metrics and Risk Assessment: The Three Tenets Model of Cybersecurity

Abstract

Progress in operational cybersecurity has been difficult to demonstrate. In spite of the considerable research and development investments made for more than 30 years, many government, industrial, financial, and consumer information systems continue to be successfully attacked and exploited on a routine basis. One of the main reasons that progress has been so meagre is that most technical cybersecurity solutions that have been proposed to-date have been point solutions that fail to address operational tradeoffs, implementation costs, and consequent adversary adaptations across the full spectrum of vulnerabilities. Furthermore, sound prescriptive security principles previously established, such as the Orange Book, have been difficult to apply given current system complexity and acquisition approaches. To address these issues, the authors have developed threat-based descriptive methodologies to more completely identify system vulnerabilities, to quantify the effectiveness of possible protections against those vulnerabilities, and to evaluate operational consequences and tradeoffs of possible protections. This article begins with a discussion of the tradeoffs among seemingly different system security properties such as confidentiality, integrity, and availability. We develop a quantitative framework for understanding these tradeoffs and the issues that arise when those security properties are all in play within an organization. Once security goals and candidate protections are identified, risk/benefit assessments can be performed using a novel multidisciplinary approach, called "QuERIES." The article ends with a threat-driven quantitative methodology, called "The Three Tenets", for identifying vulnerabilities and countermeasures in networked cyber-physical systems. The goal of this article is to offer operational guidance, based on the techniques presented here, for informed decision making about cyber-physical system security.