论文部分内容阅读
安全关键系统面临着日趋严峻的安全威胁,如何降低安全关键实时应用的风险成为挑战。通过结合风险评估与实时调度,给出一种风险驱动的任务调度方案。在分析非周期实时任务可调度性的基础上,提出一种基于处理器瞬态利用率、系统安全风险最小的在线调度近似算法。在不违背关键任务的时间约束和安全约束的前提下,该算法能够在低时间复杂度内得到安全性能确保的次优解。实验结果显示所提算法能明显降低安全关键实时应用的风险。具体来讲,最大安全风险可降低22.2%,并且在安全风险近似率上界为110%情况下,真实近似率最小为102.4%。
Security-critical systems are facing increasing security threats and how to reduce the risk of security-critical real-time applications becomes a challenge. By combining risk assessment and real-time scheduling, a risk-driven task scheduling solution is proposed. On the basis of analyzing the schedulability of non-periodic real-time tasks, an on-line scheduling algorithm based on transient processor utilization and minimum system security risk is proposed. Without compromising on mission-critical time constraints and security constraints, this algorithm can obtain a second-best solution to ensure secure performance with low time complexity. Experimental results show that the proposed algorithm can significantly reduce the risk of safety-critical real-time applications. Specifically, the maximum security risk can be reduced by 22.2%, and the minimum real approximation rate is 102.4% with the upper bound of security risk approximation of 110%.