期刊文献+

对象存储设备端数据管理策略研究 被引量:3

Research on Object Storage Device End Data Management Strategy
下载PDF
导出
摘要 在对象存储系统中,数据由智能化的对象存储设备管理.当前,对象存储设备端主要提供对象接口,传统的设计主要由通用文件I/O来封装.但是,通用文件系统在管理平坦命名空间时,尤其在长期使用后性能退化严重.因此,提出基于扩展Hash的对象文件系统XOBFS(extensible hashing objectbased storage file system),将磁盘空间划分成多个区域单元,区域中定长块分配策略与位图管理空闲块方式结合;对象属性用扩展Hash管理,在Hash桶中采用相同对象属性相邻存放策略.XOBFS应用在对象存储设备端,具有元数据规模小、长期使用性能不退化、对属性进行有效管理等特点.测试结果表明,基于大对象的XOBFS的吞吐率优于传统文件系统. With the increasing demand of high I O throughput,highly parallel data transfer and highly scalable storage system,especially in the supercomputing field,the traditional storage system has been difficult to meet the requirements.OBS(object-based storage)system is a new network storage architecture.In OBS system,local storage space allocation is managed by intelligent OSD(object-based storage device),and at present,the OSD is mainly managed by the common file system.However,when the common file system deals with the flat namespace,especially in the course of long-term use,the performance degenerates seriously.In this paper a new file system is proposed.XOBFS,which stands for extensible hashing object-based storage file system,uses fixed-length block allocation strategy and manages free block bitmap combination.Based on expanding hash manage object attribute,the same object attributes are stored in adjacent hash bucket.XOBFS applied in OSD has a lot of advantages.For example,the metadata has small scale,long-term performance is not degraded,and attribute of the object is managed effectively and so on.Test results show that big-object based XOBFS throughput rate is better than that of the traditional file system.XOBFS provides an effective method for storage management in OSD.
出处 《计算机研究与发展》 EI CSCD 北大核心 2010年第10期1832-1839,共8页 Journal of Computer Research and Development
基金 国家"九七三"重点基础研究发展计划基金项目(2004CB318201) 国家"八六三"高技术研究发展计划基金项目(2009AA01A401 2009AA01A402) 国家自然科学基金项目(60703046) 长江学者和创新团队发展计划基金项目(IRTo725)~~
关键词 对象存储系统 对象存储设备 对象文件系统 对象属性 可扩展Hash object-based storage system object-based storage device object-based file system object attribute extensible hashing
  • 相关文献

参考文献13

  • 1Gibson G A,Nagle D F,Courtright II W,et al.NASD scalable storage systems[C]//Proc of the USENIX 1999 Extreme Linux Workshop.Monterey:USENIX,1999:1-6.
  • 2Tang Hong,Gulbeden Aziz,Zhou Jingyu,et al.The Panasas active scale storage cluster-delivering scalable high bandwidth storage[C]//Proc of the 2004 ACM/IEEE Conf of Supercomputing.Los Alamitos,CA:IEEE Computer Society,2004:53-58.
  • 3Tso T Y,Tweedie S.Planned extensions to the Linux EXT2/EXT3 file system[C]//Proc of the Freenix Track:2002 USENIX Annual Technical Conf.Monterey:USENIX,2002:235-244.
  • 4Factor M,Meth K,Naor D,et al.Object storage:The future building block for storage systems[C]//Proc of the 2nd Int IEEE Symp on Mass Storage Systems and Technologies.Piscataway,NJ:IEEE,2005:119-123.
  • 5Wang Feng.File system workload analysis for large scientific computing applications[C]//Proc of the 21st IEEE/12th NASA Goddard Conf on Mass Storage Systems and Technologies.Piscataway,NJ:IEEE,2004:67-83.
  • 6Fagin R,Nievergelt J,Pippenger N,et al.Extendible hashing:A fast access method for dynamic files[J].ACM Trans on Database Systems,1979,4(3):315-344.
  • 7Heller S.Extensible hashing.Dr.Dobbs Journal of Software Tools,1989,14(11):66-70.
  • 8Wang Feng,Brandt Scott A,Miller Ethan L,et al.OBFS:A file system for object-based storage devices[C]//Proc of the 21st IEEE/12th NASA Goddard Conf on Mass Storage Systems and Technologies.Piscataway,NJ:IEEE,2004:101-118.
  • 9谢黎明,冯丹,覃灵军.对象存储系统中属性管理方法研究与实现[J].计算机研究与发展,2007,44(z1):115-121. 被引量:1
  • 10ONeil E J,ONeil P E,Weikum G.The LRU-K page replacement algorithm for database buffering[C]//Proc of ACM SIGMOD93.New York:ACM,1993:297-306.

二级参考文献11

  • 1[1]G A Gibson,D F Nagle,W Courtright Ⅱ,et al.NASD scalable storage systems.USENIX 1999 Extreme Linux Workshop,Monterey,California,1999
  • 2[2]Panasas White Paper.Object storage architecture:Defining a new generation of storage systems built on distributed,intelligent storage devices.http://www.panasas.com,2003
  • 3[3]P J Braam.The Lustre storage architecture.http://www.lustre.org/docs/lustre.pdf,2003
  • 4[4]Panasas Inc.Panasas ActiveScale Storage Cluster.http://www.panasas.com,2004
  • 5[5]D A Pease,J Menon,B Rees,et al.IBM storage Tank-a heterogeneous scalable san file system.IBM Systems Journal,2003,42(2):250-267
  • 6[6]John B Lohmeyer,George O Penokie,Paul D Aloisi,et al.Information technology-SCSI object-based storage device commands (OSD).INCITS T10 Working Draft.http://www.t10.org/drafts/htm,2004
  • 7[7]M Mesnier,G R Ganger,E Riedel.Object-based storage.IEEE Communications Magazine,2003,41(8):84-90
  • 8[8]Mike Mesnier,Gregory R Ganger,Erik Riedel.Object-based storage:pushing more functionality into storage,IEEE Potentials,2005,24(2):31-34
  • 9[9]Yingping Lu,David H C Du,Tom Ruwart.QoS provisioning framework for an OSD-based storage system.The 22nd IEEE/13th NASA Goddard Conf on Mass Storage Systems and Technologies (MSST'05),Monterey,CA,2005
  • 10[10]R Fagin,J Nievergelt,N Pippenger,et al.Extendible Hashing:A fast access method.for dynamic files.ACM Trans on Database Systems,1979,4(3):315-344

同被引文献7

引证文献3

二级引证文献1

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