计算机科学与探索 ›› 2014, Vol. 8 ›› Issue (9): 1025-1040.DOI: 10.3778/j.issn.1673-9418.1404001

• 综述·探索 • 上一篇    下一篇

云存储系统节能研究综述

于  炯1,廖  彬2+,张  陶3,孙  华1,国冰磊1,杨兴耀1   

  1. 1. 新疆大学 软件学院,乌鲁木齐 830008
    2. 新疆财经大学 统计与信息学院,乌鲁木齐 830012
    3. 新疆医科大学 医学工程技术学院,乌鲁木齐 830011
  • 出版日期:2014-09-01 发布日期:2014-09-03

A Survey on Energy-Efficient Cloud Storage System

YU Jiong1, LIAO Bin2+, ZHANG Tao3, SUN Hua1, GUO Binglei1, YANG Xingyao1   

  1. 1. School of Software, Xinjiang University, Urumqi 830008, China
    2. College of Statistics and Information, Xinjiang University of Finance and Economics, Urumqi 830012, China
    3. Department of Medical Engineering and Technology, Xinjiang Medical University, Urumqi 830011, China
  • Online:2014-09-01 Published:2014-09-03

摘要: 云存储系统作为云计算的重要组成部分,是各种云计算服务的基础。但随云存储系统规模的不断扩大和在设计时对能耗因素的忽略,使其日益暴露出高能耗、低效率的问题。因为云存储系统占整个云计算中心能耗的27%~40%,所以无论从降低服务提供商的运营成本,还是从降低能耗以保护环境的角度出发,研究云存储系统中的节能技术都具有很大的现实意义与应用前景。将存储系统中的能耗优化问题分为基于硬件的节能方法与基于调度的节能方法两大类进行讨论;并将基于调度的节能方法分为基于节点调度、基于数据调度和基于缓存预取技术3类进行综合比较;最后,对适应节能的云存储体系结构、节能模式下的QoS保证、节能模式与计算模式的匹配以及纠删码容错技术下的节能研究4个方向进行了展望。

关键词: 云计算, 云存储系统, 绿色计算, 能耗效率

Abstract: Cloud storage system as an important component of cloud computing is the foundation of all kinds of cloud computing services. However, with the expanding of its scale and the energy cost factors being ignored by its designers, the problem of high energy cost and low efficiency is exposed. Cloud storage system consumes roughly 27%~40% of the total cloud computing centers energy cost, so the research on energy saving technology of cloud storage system has great practical significance and application prospects, no matter from the point of reducing the operation cost of big data service provider or reducing energy cost to protect the environment. Firstly, this paper discusses the energy cost optimization problem of cloud storage system from two categories: energy-saving method based on hardware and energy saving method based on scheduling. Secondly, this paper divides the energy saving method based on scheduling into three aspects: the method based on node scheduling, the method based on data scheduling and the method based on cache pre-fetching technology as well, as makes a comprehensive comparison among them. Finally, this paper discusses the prospect in four directions: the cloud storage architecture which adapts to energy saving, QoS (quality of service) guarantee under energy-saving mode, the matching of energy-saving model and the calculation model, and the energy-saving research using erasure code fault-tolerant technology.

Key words: cloud computing, cloud storage system, green computing, energy-efficient