期刊文献+

基于网格存储系统的自适应数据分配策略

Method of self-adaptability data assignment on grid storage system
下载PDF
导出
摘要 在网格存储系统中通常包含各种异构的存储设备节点。基于同构系统的传统数据分配和放置策略并不能针对网格存储系统中异构存储设备的性能差异来进行数据分配,降低了网格存储系统的性能。提出一种自适应的数据分配策略(AD)。通过运用基于虚拟磁盘性能的自适应分配策略和自适应的数据副本创建策略,AD能够根据存储设备节点的性能差异有效的进行数据分配。实验证明,自适应的数据分配策略能够充分发挥异构存储设备的性能,有效解决网格存储系统的数据放置不平衡问题,提高网格存储系统的利用率。 There are usually many heterogeneous storage nodes in the grid storage system. Methods of data assignment on homogeneous storage system do not adapt to heterogeneous system any more. A new self-adaptability method (AD) of data assignment on grid storage system is proposed. It can obtain the optimal data assignment and efficiently utilize both the I/O bandwidth and storage capacity of all storage nodes. Experimental results show that this method can fully utilize the heterogeneous storage nodes' performance, resolve the imbalance problem of data placement of grid storage system and increase the utility ratio of the grid storage system effectively.
出处 《计算机工程与设计》 CSCD 北大核心 2007年第16期3918-3920,3924,共4页 Computer Engineering and Design
基金 国家十五国防预研基金项目(41501060410)
关键词 网格存储系统 异构磁盘系统 数据分配 自适应 动态分配 数据复制 grid storagesystem heterogeneous disksystem data assignment self-adaptability dynamic distribution data replication
  • 相关文献

参考文献8

  • 1刘敏,王意洁.并行I/O技术研究[J].计算机应用研究,2003,20(8):29-31. 被引量:8
  • 2Florin Isalia,Walter F View.I/O:Improving the performance of non-contiguous I/O[C].Hongkong:Proceedings of the IEEE In ternational Conference on Cluster Computer,2003:336-343.
  • 3Hwang K,Jin H Orthogonal.striping and mirroring in distributed RAID for I/O-centric cluster computing[J].IEEE Transactions on Parallel and Distributed Systems,2002,13(1):26-44.
  • 4Jose Maria Perez,Felix Garcia,Jesus Carretero,et al.Data allocation and load balancing for heterogeneous cluster storage systems[C].Tokyo,Japan:Proceedings of the 3rd IEEE/ACM International Symposium on Cluster Computing and the Grid,2003:718-723.
  • 5Zhang Jiang-lin,Fend Dan.Massive information storage[M].Beijing:Beijing Science Press,2003:161-163.
  • 6Marc Farley.Building storage network[M].2nd ed.McGraw-Hill Education,2002:237-243.
  • 7Cortes T,Labarta J.A case for heterogeneous disk arrays[R].Technical University of Catalonia,2000.
  • 8Kay R.Grid storage[EB/OL].http://www.computerworld.com/hardwaretopics/storage/story/0,10801,96674,00.html,2004.

二级参考文献14

  • 1Thakur, Rajeev. An Extended Two-phase Method for Accessing Sections of out-of-core Arrays[J]. Scientific Programming, 1996,5(4).
  • 2Madhyastha, et al. Informed Prefetching of Collective input/output Requests[ C ]. In Proceedings of SC99. November 1996.
  • 3P H Carns,et al. PVFS: A Parallel File System For Linux Clusters ( postscript, html) [ C ]. Proceedings of the 4th Annual Linux Showcase and Conference. Atlanta, GA, October 2000.317-327.
  • 4J Bester, et al. GASS: A Data Movement and Access Service for Wide Area Computing Systems. Sixth Workshop on I/O in Parallel and Distributed Systems [ C ]. May 1999.
  • 5Seamons K E. Server-directed collective I/O in Panda[EB/OL]. http : //bunny . cs . uiuc . edu/ CDR/ panda . html ,2001-09 .
  • 6Ran Oldfield, David Kotz. Armada: A Parallel File System for Computational Grids [ C ]. Proceedings of the IEEE/ACM International Symposium on Cluster Computing and the Grid.Brisbane, Australia, May 2001. 194-201.
  • 7Rajhumar Buyya. High Performance Cluster Computing [M].USA : Prentice Hall. 1999.
  • 8Cao pei. A study of Integrated Prefetching and Caching Strategies[C]. In Proceedings of the 1995 ACM SIGMETRICS, Ottawa,Canada, 1995. 188-197.
  • 9Madhyastha, et al. Reed. Exploiting Global input/output Access Pattern Classification[ C ]. IN Proceeding of SC97,1997.
  • 10Nitzberg, Bill, Virginia Lo. Collective Buffering : Improving Parallel I/O Performance [ C ]. In Proceedings of the Sixth IEEE International Symposium on High Performance Distributed Computing, August 1997.148-157.

共引文献7

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部