云环境下基于CPU密集型任务的能耗建模  

An Energy Consumption Model Based on CPU-intensive Tasks in Cloud Environment

作　　者：朱小娟[1] 雷惊鹏[1] ZHU Xiaojuan, LEI Jingpeng(Institute of Information Technology, Anhui Vocational College of Defense Technology, Liu＇ an 237000, China)

机构地区：[1]安徽国防科技职业学院信息技术学院,安徽六安237000

出　　处：《许昌学院学报》2018年第6期77-79,共3页Journal of Xuchang University

基　　金：安徽高校自然科学研究项目(KJ2018A0893);安徽高校自然科学研究项目(KJ2018ZD066)

摘　　要：在分析能耗特点的基础上,提出了非线性的能耗模型,并分别以MFCPUStresser和SPECpower_ssj2008作为CPU密集型任务的测试程序,在windows操作系统下和linux操作系统下测试了能耗及CPU利用率等指标,探讨了能耗和多个指标之间的关系.实验结果表明,预测模型的精确度能达到96%;随着CPU利用率逐渐增加,相同CPU利用率变化区间对应的能耗增加值在不断减少;且CPU密集型任务的能耗与实际内存有关,与内存使用率无关;在其它参数配置相同的情况下,内存越大,能耗越高.On the basis of analyzing the characteristics of energy consumption,a nonlinear energy consumption model is proposed. MFCPUStresser and SPECpower_ssj2008 are used as the test program of CPU-intensive task to test the energy consumption,CPU utilization and other index iunder windows OS and linux OS,respectively. The relationship between energy consumption and other indicators is discussed. The experiment results show that the accuracy of the energy consumption model can reach 96%,and the added value of energy consumption reduces continuously by the same CPU utilization change interval with the increase of CPU utilization. Power consumption of CPU intensive tasks is related to the actual memory but has nothing to do with memory usage,and the greater the memory,the higher the energy consumption in the same case as other parameters configurations.

关 键 词：能耗测试 CPU利用率 虚拟机 CPU密集型 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]