异步连接操作的峰值功率建模  

Peak power modeling for asynchronous join operations

作　　者：杨良怀[1] 张璐 范玉雷[1] YANG Lianghuai;ZHANG Lu;FAN Yulei(College of Computer Science and Technology,Zhejiang University of Technology,Hangzhou 310023,China)

机构地区：[1]浙江工业大学计算机科学与技术学院,浙江杭州310023

出　　处：《浙江工业大学学报》2018年第5期502-508,共7页Journal of Zhejiang University of Technology

基　　金：国家自然科学基金资助项目(61070042);浙江省自然科学基金资助项目(LY14F020017;LQ15F020007)

摘　　要：功率感知数据库系统的峰值功率研究是解决数据中心能效的重要议题.非运行时峰值功率的估算的挑战在于没有运行时的系统信息作为模型的输入.为克服估算困难,提出了使用CPU密集度作为CPU功耗指示量,理论上分析了异步I/O连接算法在峰值功率发生阶段的特性,通过模拟连接算法峰值功率发生阶段算法行为来估算该阶段最大CPU密集度,根据CPU密集度与CPU功率的内在联系建立异步I/O连接算法的峰值功率预测模型.实验表明:所提预测方法具有较好的预测准确性,平均相对误差低于4%.Peak power modeling in Power-aware DBMS is an important research issue in improving the energy-efficiency of data centers.The challenge of non-runtime peak power estimation lies in that there is no runtime system information for modeling.To solve this issue,using CPU-boundedness as the proxy of CPU power consumption,the characteristics of the peak power occurring stage of join algorithms are analyzed in theory.By simulating the behaviour of this stage,we estimate the maximal CPU-boundedness of join algorithms.By examining the relationship between the CPU-boundedness and CPU power,the peak power models of join algorithms under different CPU execution frequency are hence constructed.Experiments demonstrated that our proposed methods had good accuracy with the average relative error less than 4%.

关 键 词：峰值功率 功率建模 CPU密集度 数据库能效 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]