摘要
目前,高能效的并行任务调度算法设计已经成为集群系统的研究热点.现有基于复制的节能调度算法主要利用阈值平衡系统的性能和能耗,但随机设置的阈值无法根据性能需求和环境参数等特征自动调节,导致调度算法存在一定的局限性.文中提出一种面向同构集群系统的两阶段节能调度算法ATES(Adaptive Threshold-basedEnergy-efficient Scheduling).首先,设计一种基于自适应阈值的任务复制策略,该策略能够自动计算最佳阈值,利用该阈值获取近似最优的任务分组.然后,将各分组任务调度到支持DVS的处理器上,并充分利用任务之间的空闲时间降低处理器电压.该算法将任务复制策略与电压调节技术有机结合,在调度过程中能够自动调整阈值,有效提高调度算法的能效.为了验证ATES算法的合理性,通过典型应用进行仿真实验,并与常见任务调度算法进行比较,结果表明ATES算法能够更好地实现性能和能耗之间的平衡.
Increasing attention has been directly towards the energy efficient scheduling algo- rithms for parallel applications in high performance clusters. The existing duplication-based ener- gy scheduling algorithms mainly leverage a threshold to balance system performance and energy consumption. However, the threshold is given randomly which cannot flexibly adapt to the clus- ter system and application performance requirements, thus making the ideal energy efficient scheduling results. In this paper, we propose a novel two-phase Adaptive Threshold-based Energy- efficient Scheduling algorithm fATES). At first, we propose an adaptive threshold-based task duplication strategy, which can obtain an optimal tnresnom. It then leverages the optimal threshold to balance schedule lengths and energy savings by selectively replicating predecessor of a task. Therefore, the proposed task duplication strategy can get the suboptimal task groups. Then, it schedules the groups on the DVS-enabled processors to reduce processor energy whenever tasks have slack time due to task dependencies. The algorithm combines DVS(Dynamic Voltage Scal- ing) technique with adaptive threshold-based task duplication strategy. It justifies the threshold automatically to improve the energy efficiency of the scheduling algorithm. To illustrate the effec- tiveness of ATES, we simulate the real-world applications and compare ATES with the other four common task scheduling algorithms. Extensive experiment results show that our algorithm can much effectively balance schedule lengths and energy savings.
出处
《计算机学报》
EI
CSCD
北大核心
2013年第2期393-407,共15页
Chinese Journal of Computers
基金
国家自然科学基金(61202173
61272107
61103068)
高等学校博士学科点专项科研基金(20090072110035
20110072120017)
武汉大学软件工程国家重点实验室开放基金(SKLSE20080720
SKLSE2012-09-29)
同济大学嵌入式系统与服务计算教育部重点实验室开放基金资助
关键词
高性能集群
动态电压调节
任务复制
自适应阈值
调度算法
high-performance clusters
dynamic voltage scaling (DVS)
task duplication
adap tive threshold
scheduling algorithm
作者简介
刘伟,男,1978年生,博士,副教授,中国计算机学会(CCF)会员,主要研究方向为并行计算、绿色计算.E—mail:liuweiwhut@gmail.com.
尹行,男,硕士研究生,研究方向为并行计算、绿色计算.
段玉光,男,硕士研究生,研究方向为高性能计算.
杜薇,女,博士,讲师,研究方向为分布式计算、可信计算.
王伟,男,博士,副教授,中国计算机学会(CCF)会员,研究方向为绿色计算、可信计算.
曾国荪,男,博士,教授,博士生导师,中国计算机学会(CCF)高级会员,研究领域为并行计算、绿色计算.