多层次的网络服务器集群功耗管理

被引:2
作者
汤慧明
吴庆波
谭郁松
机构
[1] 国防科学技术大学计算机学院
关键词
多层次功耗管理; 自学习负载预测; 按需启动; 频率调整;
D O I
暂无
中图分类号
TP393.05 [];
学科分类号
081201 ; 1201 ;
摘要
多层次的集群功耗管理方法,在不明显影响系统性能的前提下,降低集群系统的功耗。该管理方法分为集群层次的功耗管理和本地节点层次的功耗管理。集群层次的功耗管理基于自学习负载预测的按需启动策略,根据作业的负载提供计算资源。本地节点层次的功耗管理针对负载下降产生的节点空闲问题,提出了Enhanced-conservative调控器算法,提高了负载下降时频率调整的敏感度。测试实验数据表明,该策略比其他策略能更有效的降低整个系统的功耗。
引用
收藏
页码:72 / 76
页数:5
相关论文
共 13 条
[1]  
Comparing algorithms for dynamic speedsetting of a low-power CPU. K.Govil,,E.Chan,H.Wasserman. the 1st ACM International Conference on Mobile Computing and Networking(MOBICOM-95) . 1995
[2]  
Energy-Efficient Real-T-i me Heterogeneous Server Clusters,Real-Ti me and Embedded Technology and Applications Symposium. Rusu C,Ferreira A,Scordino C,et al. Proceedings of the12th IEEE . 2006
[3]  
"Energy-aware server provisioning and load dispatching for connection-intensive internet services". Gong Chen,Wenbo He,Jie Liu,Suman Nath,Leonidas Rigas,Lin Xiao,Feng Zhao. Proceedings of the 5th USENIX Symposium on Networked Systems Design and Implementation . 2008
[4]  
Dynamic voltage scaling in multitier web servers with end-to-end delay control. Horvath,T. IEEE Transactions on Communications . 2007
[5]  
A balanced energy-effciency benchmark. Rivoire S. Proceed-ings of the2007ACM SIGMOD International Conference on Management of Data . 2007
[6]  
Scheduling for reduced CPU energy. Weiser M,Welch B,Demers A J,et al. Proceedings of the USENIX Symposium onOperating Systems Design and Implementation(OSDI’94) . 1994
[7]  
Linus about kernel governor on lkml. Torvalds L. http://marc.theaimsgroup.com/-l=linux-kernel&m=103056055008566&w=2 .
[8]  
Energy-efficient sched-uling of HPC applications in cloud computing environment. Garg S K,Yeo C S,Anandasivam A,et al. . 2009
[9]  
Optimal dynamic configuration in web server clusters. Bertini L,Leite J C B,Mosse D. . 2008
[10]  
Enhancing energy efficien-cy in multi-tier web server clusters via prioritization. Horvath T,Skadron K,Abdelzaher T. Pro-ceedings of the2007NSF Next Generation Software Work-shop,in Conjunction with the IEEE International Parallel and Distributed Processing Symposium . 2007