计算机集成制造系统 ›› 2018, Vol. 24 ›› Issue (第10): 2395-2406.DOI: 10.13196/j.cims.2018.10.003

• 当期目次 • 上一篇    下一篇

基于滑动窗口的低能耗高可靠调度算法

邓昌义1,2,郭锐锋2,3,吴昊天2,3,彭阿珍2,3,杜少华4,盖荣丽5   

  1. 1.国家工业信息安全发展研究中心系统所
    2.中国科学院大学
    3.中国科学院沈阳计算技术研究所
    4.沈阳高精数控智能技术股份有限公司
    5.大连大学信息工程学院
  • 出版日期:2018-10-31 发布日期:2018-10-31
  • 基金资助:
    国家科技重大专项资助项目(2014ZX04014-021);国家自然科学基金资助项目(61602074)。

Low-power and high-reliability scheduling algorithm based on sliding window

  • Online:2018-10-31 Published:2018-10-31
  • Supported by:
    Project supported by the National Science and Technology Major Project of the Ministry of Science and Technology,China(No.2014ZX04014-021),and the National Natural Science Foundation,China(No.61602074).

摘要: 针对开放式数控系统的高功耗和可靠性差的问题,通过优化空闲时间分配策略,实现在保证开放式数控系统可靠性前提下最小化系统能耗。提出基于滑动窗口的低能耗调度算法(LPRSW),以在保证系统可靠性的前提下降低系统功耗,同时根据任务出错场景的不同,将算法分为LPRSW-H算法和LPRSW-A算法,前者以最高速度恢复出错任务,后者采用动态电压调节后的速度恢复出错任务。在此基础上,提出一种基于滑动窗口的低能耗与可靠性协同优化调度算法,该算法对任务容错采用共享全局空闲时间代替预先给每任务都分配一个备份任务的方法,获得更多空闲时间用于降低系统功耗,同时空闲时间—能耗因子用于将空闲时间更合理地分配给后续任务,实现可靠性和低能耗之间协同优化。通过实验对比,COSALPRSW算法平均比LPRSW-A和LPRSW-H算法节约0~31.62%和0~44.4%的能耗,验证了所提方法的有效性。

关键词: 开放式数控系统, 动态电压调节, 实时调度, 能耗, 可靠性

Abstract: To solve the problem of high energy consumption and poor reliability of open CNC systems,through optimizing the slack time allocation issues,a set of real-time tasks in the open CNC system was researched to minimize energy consumption while maintaining the reliability of open CNC system.A Low Power and Reliability Based on Sliding Window (LPRSW) algorithm was proposed,which could reduce the system power consumption while ensuring the system reliability.With the different mission scenarios,LPRSW algorithm was divided into LPRSW-H and LPRSW-A algorithm,the former used the highest speed to tolerate faults,while the latter tolerated faults task by dynamic voltage scaling technology.On this basis,a Cooperative Optimal Scheduling Algorithm for Low Power and Reliability Based on Sliding Window (COSALPRSW) algorithm was built,which allocated global shared slack time to backup tasks instead of assigning a backup task to each task,and more slack time was available to reduce system power consumption.In addition,the slack-energy factor was used to rationally allocate slack time to subsequent tasks,which realized the system's reliability and low power consumption co-optimization.Experimental results showed that COSALPRSW algorithm could average 0~31.62% and 0~44.4% energy savings over LPRSW-A and LPRSW-H algorithms respectively.

Key words: open CNC system, dynamic voltage scaling, real-time scheduling, energy consumption, reliability

中图分类号: