期刊文献+

多核平台应用程序能耗分析 被引量:1

Power Character Analyze for Applications on Multi-core Platform
下载PDF
导出
摘要 多核服务器的能耗越来越巨大,当处理器核心数目增多时,能耗增加更明显。软件的动态功耗调节可以显著的降低整个系统的能耗,但这会影响到系统性能。文章分析了不同种类应用程序的能耗特性。通过硬件测量的数据,我们为软件节能调节提供了依据,得出了如下几个结论:1)软件节能有很大的空间;2)系统物理内存消耗的能耗占整体能耗;3)不同种类的应用程序,对特定硬件的节能调节可以有很好的节能效果同时可以将对系统性能的影响降到最低。 The power a multi-core processor based server consumes is enormous.This increases with the number of cores.Software based dynamic power control can efficiently reduce the power,which also can induce performance overhead for applications.In this paper,we examine the power consumption characters for different kinds of applications.Based on the analysis with the power data got from hardware,some conclusions for software dynamic power control are drawn: 1) the energy-savings by software control can be substantial;2) dram contributes more to the system power consumption compared to hard drives;3) based on the kind of applications,tuning a specific hardware's power state has little negative impact on its performance.
出处 《计算机与数字工程》 2010年第8期38-41,共4页 Computer & Digital Engineering
基金 国家自然基金重点项目(编号:60933003) 国家863计划基金项目(编号:2009AA01Z116) 陕西省科学技术研究发展计划项目(编号:2008KW-02) IBM联合实验室项目(编号:JLP200906008-1) 高效能服务器和存储技术国家重点实验室开放基金项目(编号:2009HSSA09)资助
关键词 多核 节能 性能 功耗 multicore nergy-saving performance power
  • 相关文献

参考文献12

  • 1苗蕾,齐勇,侯迪,钟虓,郑晓梅.基于遗传算法的片上多处理器任务调度策略研究[J].微电子学与计算机,2007,24(6):8-11. 被引量:6
  • 2AMD.Powernow[EB/OL].www.amd.com. 被引量:1
  • 3C.Isci,G.C.,Martonosi,M.Live.runtime phase monitoring and prediction on real systems with application to dynamic power management[C]//Proceedings of the 39th International Symposium on Microarchitecture(MICRO-39),2006,12. 被引量:1
  • 4David Cappuccio,L.C.The data center power and cooling challenge[J].Gartner,2007,11. 被引量:1
  • 5Diniz,B.,G.D.M.W.B.R.Limiting the power consumption of main memory[C]//Proceedings of the 34th International Symposium on Computer Architecture,2007. 被引量:1
  • 6Flautner,K.,Mudge,T.Vertigo:Automatic performance-setting for linux[C]//Proceedings of the Symposium on Operating Systems Design and Implementation (OSDI),2002. 被引量:1
  • 7Intel.Esit[EB/OL].www.intel.com. 被引量:1
  • 8J.Moore,J.Chase,P.R.,Sharma,R.Making scheduling cool:Temperature-aware workload placement in data centers[C]//Proceedings of the USENIX Annual Technical Conference,2005,6. 被引量:1
  • 9Miao Lei,Qi Yong,H.D.D.Y.Energy-aware scheduling tasks on chip multiprocessor[C]//Third International Conference on Natural Computation,2007,8. 被引量:1
  • 10Miao Lei,Qi Yong,H.D.D.Y.Energy saving policy for cmp system based on mmc queue with synchronous vacations[C]//Proceedings of the 2th IEEE International Conference on Innovative Computing,Information and Control,2007,9. 被引量:1

二级参考文献8

  • 1李敏强 寇纪淞 林丹.遗传算法的基本理论与应用[M].北京:科学出版社,2004.. 被引量:39
  • 2Coffman E G.Computer and job-shop scheduling theory[M].John Wiley & Sons,Inc.,1976 被引量:1
  • 3Zhong Yi Wen,Yang Jian Gang.A genetic algorithm for tasks scheduling in parallel multiprocessor systems[J].IEEE,2003,3:1785~1790 被引量:1
  • 4Hou E,Ansari N,Ren H.A genetic algorithm for multiprocessor scheduling[J].IEEE Trans Parallel and Distributed Systems,1994,5(2):113~120 被引量:1
  • 5Correa R C,Ferreira A,Rebreyend P.Scheduling multiprocessor tasks with genetic algorithms[J].1999,10(8):825~837 被引量:1
  • 6Michelle Moore.An accurate and efficient parallel genetic algorithm to schedule tasks on a cluster[J].IEEE,Proceedings of the International Parallel and Distributed Processing Symposium,2005:5 被引量:1
  • 7Krzysztof Rzadca,Franciszek Seredynski.Heterogeneous multiprocessor scheduling with differential evolution[J].IEEE 2005,3(9):2840~2847 被引量:1
  • 8钟求喜,谢涛,陈火旺.基于遗传算法的任务分配与调度[J].计算机研究与发展,2000,37(10):1197-1203. 被引量:70

共引文献5

同被引文献12

引证文献1

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部