基于负载特征聚类的节能资源调度算法

夏庆新; 兰雨晴; 唐甜; 肖利民

doi:10.13700/j.bh.1001-5965.2014.0407

基于负载特征聚类的节能资源调度算法

doi: 10.13700/j.bh.1001-5965.2014.0407

北京航空航天大学计算机学院, 北京 100191

基金项目: 国家自然科学基金重点资助项目 (61232009)

详细信息

作者简介:
夏庆新(1978—),男,辽宁辽阳人,博士生,xiaqingxin@cse.buaa.edu.cn

通讯作者:
兰雨晴(1969—),男,副教授,内蒙古呼和浩特人,lanyuqing@buaa.edu.cn,研究方向为操作系统、软件工程.

中图分类号: V221⁺.3;TB553
计量
- 文章访问数: 1077
- HTML全文浏览量: 51
- PDF下载量: 649
- 被引次数: 0
出版历程
- 收稿日期: 2014-07-09
- 修回日期: 2014-10-10
- 网络出版日期: 2015-04-20

Energy-saving resource scheduling algorithm based on workload characteristic clustering

School of Computer Science and Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100191, China

摘要

摘要: 基础设施即服务(IaaS,Infrastructure as a Service)平台提供商为用户提供高性能服务的同时,必须考虑如何在不违反服务级别协议(SLA,Service Level Agreement)的前提下,节约云平台的能耗成本.采用基于负载特征聚类的方法,提出一种IaaS云平台上保证SLA的资源调度算法,最终实现降低SLA违反率和节约能耗的目标.具体采用能耗相关的负载特征提取和改进K-means聚类分析的研究方法,进行资源调度算法研究,对物理资源进行有效分配,以保证IaaS平台节约能耗的要求.实验验证方面,通过扩展CloudSim模拟实验平台,对本研究算法与改进BFD(Best Fit Decreasing)算法进行比较,得出本研究算法在SLA违反率和节能方面更优.
- 基础设施即服务 /
- 负载 /
- 聚类 /
- 节能 /
- 服务级别协议
Abstract: When infrastructure as a service (IaaS) providers offer high performance services for users, they must think about how to reduce the energy cost of the cloud platform without violating the service level agreement (SLA). A resource scheduling algorithm to ensure SLA was proposed based on clustering analysis of the load characteristic. Ultimately, the targets of reducing SLA violation rate and saving energy were realized. The resource scheduling algorithm was analyzed based on improved K-means clustering analysis and extraction of workload characteristic according to energy consumption. Physical resources were effectively allocated to ensure the requirement of energy saving of IaaS platform. Based on the extension of the CloudSim simulation platform, the algorithm proposed was compared with the optimized best fit decreasing (BFD ) to show lower SLA violation rate and energy consumption.
- infrastructure as a service (IaaS) /
- workload /
- clustering /
- energy-saving /
- service level agreement (SLA)

HTML全文

参考文献(17)

[1]	Wang X L, Liu Z H.An energy-aware VMs placement algorithm in cloud computing environment[C]//Intelligent System Design and Engineering Application(ISDEA),2012 Second International Conference on.Piscataway,NJ:IEEE,2012:627-630.
[2]	Yahoo. M45 supercomputing project[EB/OL].Pittsburgh:Carnegie Mellon University,2009[2014-07-04].http://research.yahoo.com/node/1884.
[3]	Google. Google Cluster data V2[EB/OL].California:Googlecode,2011(2014-11-17)[2014-07-04].http://code.google.com/p/googleclusterdata/wiki/ClusterData2011.
[4]	Brown R. Report to congress on server and data center energy efficiency:public law 109-431[J].Lawrence Berkeley National Laboratory,2008:1-137.
[5]	Anil K J. Data clustering:50 years beyond K-means[J].Pattern Recognition Letters,2010,31(8):651-666.
[6]	Zhang W, Yang H I,Jiang H Y,et al.Automatic data clustering analysis of arbitrary shape with K-means and enhanced ant-based template mechanism[C]//36th IEEE Annual International Computer Software and Applications Conference.Washington,D.C.:IEEE Computer Society,2012:452-455.
[7]	孙吉贵,刘杰, 赵连宇.聚类算法研究[J].软件学报,2008,19(1):48-61. Sun J G,Liu J,Zhao L Y.Research on clustering algorithm[J].Journal of Software,2008,19(1):48-61(in Chinese).
[8]	Mahajan M, Nimborkor P,Varadarajan K.The planar K-means problem is NP-hard[J].Lecture Notes in Computer Science,2009,5431:274-285.
[9]	陈宁,陈安, 周龙骧.基于密度的增量式网格聚类算法[J].软件学报,2002,13(1):1-7. Chen N,Chen A,Zhou L X.Incremental clustering algorithm based on density[J].Journal of Software,2002,13(1):1-7(in Chinese).
[10]	Gupta H, Srivastava R.k-means based document clustering with automatic “k” selection and cluster refinement[J].International Journal of Computer Science and Mobile Applications,2014,2(5):7-13.
[11]	李凯,李昆仑, 崔丽娟.模型聚类及在集成学习中的应用研究[J].计算机研究与发展,2007,44(z2):203-207. Li K,Li K L,Cui L J.Study of model clustering and its application to ensemble learning[J].Journal of Computer Research and Development,2007,44(z2):203-207(in Chinese).
[12]	Polczynski M, Polczynski M.Using the k-means clustering algorithm to classify features for choropleth maps[J].Cartographica:The International Journal for Geographic Information and Geovisualization,2014,49(1):69-75.
[13]	Coffman Jr E G, Csirik J,Galambos G,et al.Bin packing approximation algorithms:survey and classification[M].Handbook of Combinatorial Optimization.New York:Springer,2013:455-531.
[14]	Hong S, Zhang D,Lau H C,et al.A hybrid heuristic algorithm for the 2D variable-sized bin packing problem[J].European Journal of Operational Research,2014,238(1):95-103.
[15]	Wang L, Lan Y Q,Xia Q X.Using CloudSim to model and simulate cloud computing environment[C]//Computational Intelligence and Security(CIS),2013 9th International Conference on.Emeishan:IEEE,2013:323-328.
[16]	SPEC. SPECpower_ssj2008[EB/OL].Texas:Standard Performance Evaluation Corporation,2008(2013-3-21)[2014-07-04].http://www.spec.org/power/docs/SPECpower-Device_List.html.
[17]	Beloglazov A, Buyya R.Optimal online deterministic algorithms and adaptive heuristics for energy and performance efficient dynamic consolidation of virtual machines in cloud data centers[J].Concurrency and Computation:Practice and Experience,2012,24(13):1397-1420.