面向高性能计算的混合存储系统设计与实现

上海交通大学学报（自然版） ›› 2013, Vol. 47 ›› Issue (01): 113-117.

• 自动化技术、计算机技术 • 上一篇下一篇

面向高性能计算的混合存储系统设计与实现

宋振龙, 李琼, 徐炜瑕, 李晋文, 刘光明

(国防科学技术大学计算机学院, 长沙 410073)

收稿日期:2012-05-30 出版日期:2013-01-30 发布日期:2013-01-30
基金资助:
国家高技术研究发展计划(863)项目(2012AA01A301)，国家自然科学基金项目（60873212）资助

Realization and Design of a Hybrid Storage System in High Performance Computing

SONG Zhen-Long, LI Qiong, XU Wei-Xia, LI Jin-Wen, LIU Guang-Ming

（College of Computer, National University of Defense Technology, Changsha 410073, China）

Received:2012-05-30 Online:2013-01-30 Published:2013-01-30

摘要/Abstract

摘要： 提出了一种采用RAM-disk的混合存储系统，在计算节点的并行文件系统的元数据服务器和对象存储服务器组件中设置不同用途的RAM-disk，用于缓存各种访问模式的文件并提高系统的读写性能.结果表明，采用RAM-disk的混合存储系统可以明显提高系统的读写性能，同时可提高系统的可用性和可维性.

关键词: Lustre文件系统, I/O性能, 高速缓冲存储器, 混合存储系统

Abstract: A hybrid storage system with RAM-disk is proposed. It was a novel method which can reduce the latency of I/O by inserting a RAM-disk in the different nodes of the system for caching the I/O files. The result of experiment implies that the hybrid storage system with RAM-disk can improve the performance and availability of file system.

Key words: Lustre file system, I/O performance, Cache, hybrid storage system

中图分类号:

TP 333

宋振龙, 李琼, 徐炜瑕, 李晋文, 刘光明. 面向高性能计算的混合存储系统设计与实现[J]. 上海交通大学学报（自然版）, 2013, 47(01): 113-117.

SONG Zhen-Long, LI Qiong, XU Wei-Xia, LI Jin-Wen, LIU Guang-Ming. Realization and Design of a Hybrid Storage System in High Performance Computing[J]. Journal of Shanghai Jiaotong University, 2013, 47(01): 113-117.

参考文献

［1］Braam P J. The lustre storage architecture ［EB/OL］. ［20120601］. http:∥ www.clusterfs.com.
［2］Nagle D, Serenyi D, Matthews A . The panasas activescale storage cluster: Delivering scalable high bandwidth storage ［C］ ∥ Proceedings of the ACM/IEEE SC2004 Conference on Supercomputing. Washington: IEEE Computer Society, 2004: 5362.
［3］Rodeh O, Teperman A. zFSa scalable distributed file system using object disks ［C］ ∥ Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies. Washington: IEEE Computer Society, 2003: 207218.
［4］Menon J, Pease D A, Rees R, et al. IBM storage tank: A heterogeneous scalable SAN file system ［J］. IBM Systems Journal, 2003, 42(2): 250267.
［5］Bryant R, Raddatz D, Sunshine R. PenguinoMeter: A new FileI/O Benchmark for Linux ［C］∥ Proceedings of the 5th Annual Linux Showcase and Conference. Oakland, CA: USENIX, 2001: 510.
［6］Griffioen J, Appleton1 R. The design, implementation, and evaluation of a predictive caching file system ［R］. USA：Department of Computer Science, University of Kentucky: 1996.
［7］Roselli D, Lorch J R, Anderson T E. A comparison of file system workloads ［C］∥ Proceedings of the 2000 USENIX Annual Technical Conference. San Diego, USA: USENIX, 2000.
［8］Ruemmler C, Wilkes1 J. An introduction to disk drive modeling ［J］. IEEE Computer, 1994, 27(3): 17.
［9］李柱, 周恩强, 廖湘科. Filter Cache：一种提高Lustre I/O性能的方法［J］. 计算机研究与发展, 2009, 46 (z2): 7177.
LI Zhu, ZHOU Enqiang, LIAO Xiangke. Filter Cache: A method for improving I/O performance of lustre file system ［J］. Journal of Computer Research and Development, 2009, 46(z2): 7177.
［10］刘光明, 邹丹, 张超. 面向Lustre文件系统的固态盘存储加速技术研究［J］.计算机研究与发展, 2009, 46(z2): 371375.
LIU Guangming, ZOU Dan, ZHANG Chao. Research on lustreoriented storage acceleration with solid state disk ［J］. Journal of Computer Research and Development, 2009,46(z2): 371375.
［11］李林林, 吴万国, 孙良旭. 并行文件系统Lustre细粒度I/O性能优化［J］. 计算机工程与应用, 2012, 48(17): 8892.
LI Linlin, WU Wanguo, SUN Liangxu. Performance of finegrained I/O in parallel file system of Luster ［J］. Computer Engineering and Applications, 2012 48(17): 8892.
［12］Wang F. Storage management in large distributed objectbased storage system ［M］. USA: University of California Santa Cruz, 2006.

[1]	孙岩, 宋超, 黎铁军, 张民选. 一种基于三维堆叠技术的高可靠性Cache结构[J]. 上海交通大学学报（自然版）, 2013, 47(01): 65-69.
[2]	张建民, 张峻, 夏军, 庞征斌, 徐炜遐. 一种高速缓冲存储器的可综合伪随机功能验证方法[J]. 上海交通大学学报（自然版）, 2013, 47(01): 123-128.

面向高性能计算的混合存储系统设计与实现

Realization and Design of a Hybrid Storage System in High Performance Computing

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价