新的混合关键任务调度算法的研究

New Scheduling Algorithm for Mixed-Criticality Real-Time Task Sets

  • 摘要: 分析了混合关键系统中当前任务调度方法存在的问题, 提出一种正反向时间分割和关键因子优先的调度算法, 该方法将所有混合关键任务按照不同级别进行时间正反向分割得出空闲时间窗口, 并根据关键因子的大小决定优先级别. 为了减少在关键级别转化过程中优先级低的任务丢失死限的工作数量, 将空闲窗口分配给由于关键级别转化而使得优先级落后的任务. 仿真实验表明, 该方法在降低任务丢失死限率和完成任务的数量方面比按照关键级别进行优先级指派算法(CAPA)和OCBP方法较优.

     

    Abstract: By analyzing the problems of real-time scheduling in mix-criticality systems, a new scheduling algorithm, forward and backward time window partition-criticality factor prior (FBTWP-CFP), is proposed. It can separate the running time windows for the tasks of all criticality levels offline from the forward direction and backward direction and generate the idle windows for the tasks which get the lower priority by the criticality factor for the criticality level changed. The simulation results show that FBTWP-CFP is better than criticality assigned priority algorithm (CAPA) and own criticality based priority (OCBP) in completed tasks number and reduced miss deadline ratio.

     

/

返回文章
返回