组合盘中结合访问次数与能量代价的文件缓存替换策略
作者:
基金项目:

国家自然科学基金(61070042)以及浙江省自然科学基金(Y1090096)


Frequency and Energy-Based Replacement Scheme for Heterogeneous Drive
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [11]
  • |
  • 相似文献
  • | | |
  • 文章评论
    摘要:

    如何有效地降低存储子系统能耗是近几年研究的热点议题。新型非易失、抗震、低功耗闪存及固态盘的出现给存储子系统节能带来了新的机会。但其每单位价格昂贵,目前难以替代硬盘的角色。结合硬盘和固态盘的优势,本文采用组合盘(由硬盘和固态盘组成)节能。结合文件访问次数和能量代价,我们提出了改进的文件缓存替换策略FEBR(Frequency & Energy-based replacement)。实验采用两个真实办公用户数据,结果表明组合盘方案是可行的,节能百分比可达70%~80%;与经典替换算法、最新较好的ARC 算法以

    Abstract:

    How to reduce the energy consumption of storage sub-system effectively has gained a lot of attention these years. The emergence of non-volatile, shock resistant and low power flash memory and solid state driver (SSD) brings a new opportunity for power-saving in storage sub-system. However, it still cannot totally replace hard disk for its higher cost per storage unit. This paper focuses on the heterogeneous drive scheme (heter-Drive for short) for both energy-conservation. By considering both file access frequencies and energy cost, we propose an improved file cache replacement scheme called FEBR (Frequency and Energy-based Replacement). We collected four real-world office users' file access data for experiments. The results show that heter-Drive does works well and saves as high as 70% ~ 80% of energy; the extensive comparisons with the classic replacement algorithms, the widely used ARC and the optimal algorithm OPT show that FEBR is consistently better than other alternatives.

    参考文献
    1 Douglis F, Krishnan P, Marsh B. Thwarting the power hungry disk. USENIX Winter Conference. 1994,292-306.
    2 Greenawalt P. Modeling power management for hard disks. Workshop on Modeling, Analysis, and Simulation on Computer and Telecommunication Systems, 1994,62-66.
    3 Deng Y. What is the future of disk drives, death or rebirth?. ACM Computing Surveys, 2011,43(3).
    4 Intel X18-M/X25-M SATA Solid State Drive Product Manual.2009, May. http://download.intel.com/design/flash/nand/mainstream/ mainstream-sata-ssd-datasheet.pdf.
    5 Narayanan D, Thereska E, et.al. Migrating server storage to SSDs: analysis of tradeoffs. 2009, EuroSys.
    6 Liu S, Cheng X, Guan X, Tong D. Energy efficient management scheme for heterogeneous secondary storage system in mobile computers. ACM Symposium on Applied Computing,2010,251-257.
    7 Belady LA. A study of replacement algorithms for virtual storage computers. IBM Systems Journal, 1966,5(2):78-101.
    8 Mattson RL, Gecsei J, Slutz DR, Traiger IL. Evaluation Techniques for Storage Hierarchies. IBM Systems Journal,1970,9(2):78-117.
    9 Megiddo N, Modha DS. ARC: A Self-Tuning, Low Overhead Replacement Cache. USENIX Conference on File and Storage Technologies (FAST), 2003,115-130.
    10 Robinson JT, Devarakonda MV. Data cache management using frequency-based replacement. ACM SIGMETRICS Conference. 1990.134-142.
    11 Li K, Kumpf R, Horton P, Anderson T. A Quantitative Analysis of Disk Drive Power Management in Portable Computers. USENIX Winter Conference. 1994.279-292.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

周健,杨良怀,龚卫华.组合盘中结合访问次数与能量代价的文件缓存替换策略.计算机系统应用,2012,21(6):208-211,207

复制
分享
文章指标
  • 点击次数:2025
  • 下载次数: 3583
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2011-09-16
  • 最后修改日期:2011-10-29
文章二维码
您是第12478933位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号