Disk power module and evaluation based on I/O idle rate

Computer Engineering and Applications ›› 2014, Vol. 50 ›› Issue (10): 216-220.

Previous Articles Next Articles

Disk power module and evaluation based on I/O idle rate

ZHANG Haipeng1, LI Xi1，2, CHEN Xianglan1, ZHU Zongwei1，2

1.School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China
2.Suzhou Institute for Advanced Study, University of Science and Technology of China, Suzhou, Jiangsu 215123, China

Online:2014-05-15 Published:2014-05-14

基于I/O空闲率的硬盘功耗建模与评估方法研究

张海鹏1，李曦1，2，陈香兰1，朱宗卫1，2

1.中国科学技术大学计算机科学与技术学院，合肥 230027
2.中国科学技术大学苏州研究院，江苏苏州 215123

Abstract

Abstract: Storage is consuming more and more energy in computer, especially in data center. In order to reduce disk power, it needs to establish the disk power module when there no hardware to measure the power. Modern operating system uses file systems to manage I/O and provide real-time monitoring data. This paper analyzes the I/O processing of EXT2, EXT4 and NILFS2 file systems, and discovers that different file systems have different influence on disk power. This paper presents I/O idle rate to depict the difference, more specifically, proposes the disk power module and evaluates method by using I/O idle rate. This paper also uses I/O idle rate to reduce disk power, when improving the I/O idle rate of video from 4.79% to 96.55, reducing the disk power 45.28%.

Key words: file system, power module, I/O idle rate, power optimization

摘要： 存储设备的功耗在整个计算机系统尤其是数据中心所占的比例已经越来越大。要进行硬盘功耗优化，在缺少硬件直接测量硬盘功耗的情况下就需要对硬盘功耗进行建模。现代操作系统通过文件系统层进行硬盘I/O操作管理，并提供实时监控数据。通过对EXT2、EXT4和NILFS2等常用文件系统的硬盘I/O过程进行实验分析，发现不同的文件系统对硬盘功耗具有不同的影响，为了刻画这种差异性，提出基于I/O空闲率的硬盘功耗建模与评估方法。同时使用I/O空闲率来指导硬盘进行功耗优化，将视频播放器的I/O空闲率由4.79%提高到96.55%，硬盘的功耗降低45.28%。

关键词: 文件系统, 功耗模型, I/O空闲率, 功耗优化

ZHANG Haipeng1, LI Xi1，2, CHEN Xianglan1, ZHU Zongwei1，2. Disk power module and evaluation based on I/O idle rate[J]. Computer Engineering and Applications, 2014, 50(10): 216-220.

张海鹏1，李曦1，2，陈香兰1，朱宗卫1，2. 基于I/O空闲率的硬盘功耗建模与评估方法研究[J]. 计算机工程与应用, 2014, 50(10): 216-220.

[1]	XU Qi1，2, WANG Cong1，2, CHENG Yaodong1, CHEN Gang1. Cross-Domain File System for Distributed Sites [J]. Computer Engineering and Applications, 2019, 55(8): 1-8.
[2]	WU Yaoyao1, YANG Geng1，2. Distributed File System Load Balancing in Cloud Environment [J]. Computer Engineering and Applications, 2019, 55(10): 67-72.
[3]	MA Zhen, HALIDAN Abudureyimu, LI Xitong. Research on access optimization of small files in massive sample data sets [J]. Computer Engineering and Applications, 2018, 54(22): 80-84.
[4]	GUO Kai1, GONG Caixin1, GONG Yili1, LEI Yingchun2. Research on scan scheduling in MPP databases on distributed file systems [J]. Computer Engineering and Applications, 2018, 54(13): 84-87.
[5]	LIU Haibo1, DONG Yujie2. Design and implementation of error-correcting code storage based on MooseFS [J]. Computer Engineering and Applications, 2017, 53(6): 91-95.
[6]	MIAO Xiaolong1, CHEN Hao1, ZHONG Jiang2. Energy-conserving strategies of file storage based on cluster scale adjustment [J]. Computer Engineering and Applications, 2017, 53(24): 80-85.
[7]	LIU Shuoyang, ZHOU Lijuan, REN Zhongshan, ZHANG Shudong. HDFS load balancing in ophthalmic medical image file access [J]. Computer Engineering and Applications, 2017, 53(2): 253-259.
[8]	LI Qiang1，2, SUN Zhenyu1，2, SUN Gongxing1. Data random access method oriented to HDFS [J]. Computer Engineering and Applications, 2017, 53(10): 1-7.
[9]	LIN Sui, HUANG Jian, JIANG Wenchao. Co-Work: Data consistency preserving algorithm based on HDFS in secure cloud storage system [J]. Computer Engineering and Applications, 2017, 53(1): 9-15.
[10]	LI Sanmiao, LI Longshu. Performance analysis of four methods for handling small files in Hadoop [J]. Computer Engineering and Applications, 2016, 52(9): 44-49.
[11]	NIE Ruihua1, XIE Wenjun1, LIANG Jun2. Improved I/O policy of small files based on cache pool [J]. Computer Engineering and Applications, 2016, 52(16): 210-215.
[12]	CHEN Dejun1, GAO Xiaojun1, WANG Yifei2. Load balancing study in cloud storage based on AHP [J]. Computer Engineering and Applications, 2015, 51(7): 56-60.
[13]	XUE Shishuai, LIU Dan, XU Zhan, LI Jianhong. Design of secure file system for active RFID tag [J]. Computer Engineering and Applications, 2014, 50(24): 47-49.
[14]	YUAN Yu1，2, CUI Chaoyuan2, WU Yun2,CHEN Zhuhong2，3. Performance analysis of Hadoop for handling small files in single node [J]. Computer Engineering and Applications, 2013, 49(3): 57-60.
[15]	WANG Dong1, WANG Chao2，3, HAN Yong2, XIAO Zhuo2. Protocol overhead comparison of iSCSI and NFS [J]. Computer Engineering and Applications, 2012, 48(36): 105-111.

Disk power module and evaluation based on I/O idle rate

基于I/O空闲率的硬盘功耗建模与评估方法研究

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics