【作者】 张宁；

【导师】 徐成；

【作者基本信息】 湖南大学 ， 计算机科学与技术， 2014， 硕士

【摘要】 在安全性关键(Safety-Critical)实时嵌入式系统(航空电子系统、汽车电子系统等)中,随着功能数量在不断增加,系统结构也在变得越来越复杂。为了达到性能提高、成本节约、多样化的功能和安全可靠的效果,系统将具有不同关键级的功能模块集成到一个共享硬件上。对于这样的混合关键级系统,在同一个硬件资源下对不同关键级的任务进行调度将会出现任务之间的互相影响和干扰,会对这些任务的调度带来很多新的挑战和问题。当前基于响应时间分析的混合关键级系统调度算法研究存在响应时间分析不充分、任务调度模型不完善等方面的不足。本文针对这些问题,对基于响应时间分析的混合关键级系统实时调度算法进行研究。本文主要研究工作如下:鉴于当前基于响应时间分析的悲观周期混合关键级任务调度算法中的响应时间分析算法AMC(Adaptive Mixed Criticality)不够精确,主要是在高关键级任务对任务抢占时间的估计存在冗余,本文对现有响应时间分析算法AMC等进行详细分析,并对AMC进行改进从而减少高关键级任务对任务抢占时间的冗余估计,使之计算出的任务响应时间更精确,然后在此基础上利用最佳优先级分配算法MC-OPA(Mixed-Criticality-Optimal Priority Assignment)来分配任务的优先级,并结合固定优先级调度策略,提出了基于改进响应时间分析(SAMC)的悲观周期混合关键级任务调度算法。在实验中通过与目前基于响应时间分析的悲观周期混合关键级任务调度算法进行性能对比,来验证该算法改进的有效性。针对现有基于响应时间分析的悲观周期混合关键级任务调度算法研究大都只考虑双关键级的情况,无法很好地应用到关键级数大于两个关键级的多关键级系统中,本文对任务在系统运行过程中多个关键级状态下的响应时间进行详细分析,并对现有的响应时间分析算法(AMC)和改进的响应时间分析算法(SAMC)进行扩展使之支持多关键级的情况,从而能很好的适用于具有悲观周期的混合多关键级系统。然后在此基础上利用MC-OPA算法来分配任务的优先级,并结合固定优先级调度策略,得到基于时间响应分析的支持多关键级的悲观周期混合关键级任务调度算法。更多还原

【Abstract】 In modern safety-critical real-time embedded systems(such as automotive electronic systems,avionics systems),along with the increasing number of applications,the structure of the systems are more and more complex,considering performance,cost savings,functional diversification,safety and reliability,an increasingly important trend of these systems is the integration of components with different levels of criticality onto a common hardware platform.In such a "mixed-criticality" system,different tasks with different criticalities may have interference and influence on each other,which would bring a lot of new challenges and scheduling problems..In the current researches on mixed-criticality system scheduling area,there exist a lot of deficiencies such as deficient algorithm design,inadequate scheduling model.In order to solve these problems,we do researches on the mixed-criticality real-time scheduling algorithm based on response-time analysis.The main research works are as follows:Considering the existing response-time analysis AMC(Adaptive Mixed Criticality)for the tasks with pessimistic period in the mixed-criticality real-time scheduling algorithm are not enough precise and sufficient that it exists the excessive high-criticality tasks’ interference on the response time of a task,this paper analyses and introduces the existing response-time analysises for tasks with pessimistic period and makes an improvement on AMC,which reduces the high-criticality tasks’ interference on the response time of a task,in order to make it more precise.Based on it we present a sufficient adaptive response-time analysis for mixed-criticality systems(SAMC)considering pessimistic period parameters and MC-OPA(Mixed-Criticality-Optimal Priority Assignment)scheme for fixed-priority uniprocessor scheduling of mixed-criticality system with dual-criticality levels.In the experiments,this paper illustrates the effectiveness of the algorithm by comparing its performance with other algorithms’.Taking into account that the most existing mixed-criticality scheduling algorithms based on response-time analysis for tasks with pessimisitic period only consider the case of dual-criticality systems which can not be suitable to the mixed-criticality system having more than two criticalities,through the detailed analysis of the task’s response time in the several criticality levels this paper extends the existing response-time analysis(AMC)and improved response-time analysis(SAMC)for dual-criticality systems to multi-criticality systems in order to make them suitable to multi-criticality systems with pessimisitic period.Then we present two response-time analysis schemes for mixed-criticality systems considering pessimistic period parameters and MC-OPA(Mixed-Criticality-Optimal Priority Assignment)scheme for fixed-priority uniprocessor scheduling of mixed-criticality system with multi-criticality levels.更多还原