电子学报 ›› 2016, Vol. 44 ›› Issue (4): 893-897.DOI: 10.3969/j.issn.0372-2112.2016.04.020

• 学术论文 • 上一篇    下一篇

面向异构计算的能效感知调度研究

王静莲1,2, 龚斌1,3   

  1. 1. 山东大学计算机科学与技术学院, 山东济南 250101;
    2. 鲁东大学信息与电气工程学院, 山东烟台 264025;
    3. 山东省高性能计算中心, 山东济南 250101
  • 收稿日期:2014-10-23 修回日期:2014-12-18 出版日期:2016-04-25
    • 作者简介:
    • 王静莲 女,1979年5月出生于山东省莱州市.现为山东大学博士研究生.主要从事并行与高性能计算、绿色计算和多目标全局优化算法等方向的研究. E-mail:wjljing@163.com;龚 斌 男,1964年10月生于山东省济南市.教授、博士生导师.现为山东大学计算中心主任、山东省高性能计算中心副主任,主要从事网格与高性能计算、机群计算方面的研究工作. E-mail:gb@sdu.edu.cn
    • 基金资助:
    • 国家自然科学基金 (No.61272094); 国家863高技术研究发展计划 (No.2006AA01A113,No.2012AA01A306)

Research on Energy-Efficiency Aware Heterogeneous Scheduling

WANG Jing-lian1,2, GONG Bin1,3   

  1. 1. College of Computer Science and Technology, Shandong University, Jinan, Shandong 250101, China;
    2. College of Information and Electrical Engineering, Ludong University, Yantai, Shandong 264025, China;
    3. Shandong High Performance Computing Center, Jinan, Shandong 250101, China
  • Received:2014-10-23 Revised:2014-12-18 Online:2016-04-25 Published:2016-04-25
    • Supported by:
    • National Natural Science Foundation of China (No.61272094); National High Technology Research and Development Program of China  (863 Program) (No.2006AA01A113, No.2012AA01A306)

摘要:

异构调度可使大规模计算系统采用并行方式聚合广域分布的各种资源以提高性能.传统调度目标追时限约束求高性能而忽视高效能,远不能适应绿色计算科学发展要求.因此,本文在理论上一方面建立融合能效感知的调度模型;另一方面提出适于超计算机混合体系的多学科背景的元启发式优化算法.从技术上解决了面向不同环境目标的调度实施条件界定及调度指标(时间、能耗)实时变化描述等问题.大量仿真实验结果表明:与三个元启发式调度器相比,论文方法在能效及可扩展等方面优势明显;对于高维实例,整体性能改善分别达到8%,15%和17%.

关键词: 异构调度, 绿色计算, 协同进化, 混合并行

Abstract:

Enabled to provide pervasive access to distributed resources in parallel ways,heterogeneous scheduling is extensively applied in large-scaled computing system for high performance.Conventional real-time scheduling algorithms,however,disregard energy-efficiency in addition to stringent timing constraints.In recognition of green computing,an energy-aware model is firstly presented.Secondly,inspired by multi disciplines,the meta-heuristic is addressed based on the supercomputer hybrid architecture.On the other hand,some technological breakthroughs are achieved,including boundary conditions for different heterogeneous computing and grid scheduling and descriptions of real-time variation of scheduling indexes (stringent timing constraints and energy-efficiency).Extensive simulator and simulation experiments highlight higher efficacy and better scalability for the proposed approaches compared with the other three meta-heuristics;the overall improvements achieve 8%,15% and 17% for high-dimension instances,respectively.

Key words: heterogeneous scheduling, green computing, co-evolution, hierarchical parallelization

中图分类号: