| 面向动态异构多核处理器的公平调度算法 |
| |
| 引用本文： | 王涛,安虹,孙涛,高晓川,张海博,程亦超,彭毅.面向动态异构多核处理器的公平调度算法[J].软件学报,2014,25(S2):80-89. |
| |
| 作者姓名： | 王涛 安虹 孙涛 高晓川 张海博 程亦超 彭毅 |
| |
| 作者单位： | 中国科学技术大学 计算机科学与技术学院, 安徽 合肥 230027,中国科学技术大学 计算机科学与技术学院, 安徽 合肥 230027,中国科学技术大学 计算机科学与技术学院, 安徽 合肥 230027,中国科学技术大学 计算机科学与技术学院, 安徽 合肥 230027,中国科学技术大学 计算机科学与技术学院, 安徽 合肥 230027,中国科学技术大学 计算机科学与技术学院, 安徽 合肥 230027,中国科学技术大学 计算机科学与技术学院, 安徽 合肥 230027 |
| |
| 基金项目： | 国家自然科学基金(60970023);国家重点基础研究发展计划(973)(2011CB302501);国家高技术研究发展计划(863)(2012AA010902,2012AA010901) |
| |
| 摘 要： | 动态异构多核处理器的处理器核可动态调整的特征给操作系统调度算法带来了新的机遇和挑战.利用处理器核动态可调整的特征能更好地适应不同任务的运行需求,带来巨大的性能优化空间.然而也带来新的代价和更复杂的公平性的计算.为了解决面向动态异构多核处理器结构上的公平性调度问题,提出了一个基于集中式运行队列的调度模型,以降低调度算法在动态处理器核变化所带来的维护开销.并重新思考在动态异构处理器结构下公平性的定义,基于原有CFS调度算法提出新的HFS调度算法.HFS调度算法不仅能简单而有效地利用动态异构多核处理器的性能优势,而且能提供在动态异构多核处理器上的公平性调度.通过模拟SCMP,ACMP,DHCMP平台,证明了提出的HFS调度算法能够很好地发挥DHCMP结构的性能特征,比运行目前主流调度算法的SCMP和ACMP结构提升10.55%的用户级性能(ANTT),14.24%的系统吞吐率(WSU).
|
| 关 键 词： | 动态异构多核 集中式运行队列 任务调度 资源分配 公平性算法 |
| 收稿时间： | 8/5/2013 12:00:00 AM |
| 修稿时间： | 2014/3/13 0:00:00 |
Fair Scheduling on Dynamic Heterogeneous Chip Multiprocessor |
| |
| WANG Tao,AN Hong,SUN Tao,GAO Xiao-Chuan,ZHANG Hai-Bo,CHENG Yi-Chao and PENG Yi.Fair Scheduling on Dynamic Heterogeneous Chip Multiprocessor[J].Journal of Software,2014,25(S2):80-89. |
| |
| Authors: | WANG Tao AN Hong SUN Tao GAO Xiao-Chuan ZHANG Hai-Bo CHENG Yi-Chao and PENG Yi |
| |
| Affiliation: | School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China,School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China,School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China,School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China,School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China,School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China and School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China |
| |
| Abstract: | Dynamic Heterogeneous CMPs (DHCMP), which provide the capability to configure different number and types of processing cores at system runtime, dramatically improve energy- and power-efficiency by scheduling workloads on the most appropriate core type. A significant body of recent work has focused on improving system throughput through scheduling on asymmetric CMPs (ACMP). However, none of the prior work has looked into fairness. In this work, centralized run queue is introduced and a heterogeneity-aware fair scheduler (HFS) is proposed to address the fair scheduling problem on DHCMP. HFS algorithm can not only gain the capability of DHCMP to configure the types of processing cores to match the granularities of parallelism in the tasks, but also keep the fairness when tasks running simultaneously. Experimental results demonstrate that HFS on DHCMP outperforms the best performing fair scheduler on SCMP and ACMP by 10.55% in user-oriented performance (ANTT), and 14.24% in system throughput (WSU). |
| |
| Keywords: | dynamic heterogeneous centralized run queue task scheduler resource allocation fair algorithm |
|
| 点击此处可从《软件学报》浏览原始摘要信息 |
|
点击此处可从《软件学报》下载全文 |
|
