摘要 : Reconfiguration can be used to improve system performance, increase fault-tolerance, recover from failures, prevent and recover from cyberattacks. The process of using reconfiguration to increase a system's resilience to cyberatta... 展开 Reconfiguration can be used to improve system performance, increase fault-tolerance, recover from failures, prevent and recover from cyberattacks. The process of using reconfiguration to increase a system's resilience to cyberattacks is called Moving Target Defense (MTD). While system reconfiguration has many advantages, it brings potential disadvantages such as performance and availability degradation when a system is undergoing reconfiguration. This paper analyzes the use of MTD to reduce the frequency of cyberattacks and develops a closed-form analytic model that captures the tradeoffs between resilience to cyberattacks and system performance. The state-of-the art in MTD discusses how proposed methods operate and uses experimentation to demonstrate their use. This paper fills a research gap in terms of developing formal models that analyze tradeoffs between performance and resilience to cyberattacks. More specifically, the main contributions of this paper include: (1) A closed-form analytic model for computing the probability that an attacker is successful in completing the reconnaissance phase of the attack and the computation of the task's execution time when regular reconfigurations are used. (2) An optimization model for determining the optimal task reconfiguration frequency in a way that takes into account given performance-security tradeoffs. (3) A closed-form analytic model for computing the probability that an attacker is successful in completing the reconnaissance phase of the attack and the computation of the task's execution time when irregular reconfigurations are used. (4) Derivation of the maximum probability that an attacker fails in completing the reconnaissance phase under irregular reconfiguration intervals. 收起