基于概率性分析的混合关键级系统节能调度算法

Energy Saving Scheduling Algorithm of Mixed-Criticality System Based on Probability Analysis

针对混合关键级系统中的固定优先级任务节能问题,文中提出了基于概率性分析的混合关键级系统节能调度算法。混合关键级系统的实时性要求使得系统建模和分析偏向于较坏的情况。该类系统中出现任务超限的情况相对较少,易存在资源配置过度问题。通过DVFS(Dynamic Voltage Frequency Scaling)技术和混合关键级系统调度算法相结合的方式挖掘空闲时间,从而在保证系统实时性的前提下降低系统的能耗。利用MCSIMU仿真软件对所提算法进行了仿真验证,实验结果表明,对于固定优先级任务与未使用节能调度算法相比,固定优先级节能调度算法的节能率可达45%。

In view of the problem of energy saving of fixed priority tasks in mixed criticality system,a probabilistic analysis based energy-saving scheduling algorithm for mixed criticality system is proposed in this study.The real-time requirement of hybrid critical system makes the system modeling and analysis biased to the worse case.In this type of system,task overlimit is relatively rare,and resource overallocation is easy to occur.In this study,DVFS(Dynamic Voltage Frequency Scaling)technology and mixed criticality system scheduling algorithm are combined to mine idle time,so as to reduce the energy consumption of the system on the premise of ensuring the real-time performance of the system.The proposed algorithm is verified by MCSIMU(Mixed Criticality System Simulation)simulation software.The experimental results show that the energy saving rate of the fixed priority energy-saving scheduling algorithm can reach 45%compared with that of the unused energy-saving scheduling algorithm.