A Metadata Management Method Based on Directory Path
Affiliation:

  • Article
  • | |
  • Metrics
  • |
  • Reference [12]
  • |
  • Related [20]
  • |
  • Cited by [7]
  • | |
  • Comments
    Abstract:

    A metadata management method dividing directory path attribute from directory object is proposed, which extends the present object storage architecture. This method avoids efficiently the large-scale metadata migration according to the updating directory attributes, improves the cache utilization and hit rate by reducing the overlap cache of prefix directory, reduces the disks I/O demands by reducing the overhead of traversing the directory path and exploiting directory locality, and avoids overloading a single metadata server by dynamic load balancing. Experimental results demonstrate that this method has obvious advantages in improving the throughput, scalability, balancing metadata distribution, and in reducing the metadata migration.

    Reference
    [1]Nagle D,Serenyi D,Matthews A.The Panasas ActiveScale storage cluster delivering scalable high bandwidth storage.In:Benton V,ed.Proc.of the ACM/IEEE SC 2004 Conf.Washington:IEEE Computer Society,2004.53-62.
    [2]Lustre SP.Building a file system for 1000 node clusters.In:John WL,ed.Proc.of the 2003 Ottawa Linux Symp.Ottawa:Red Hat,Inc.,2003.401-407.
    [3]Brandt SA,Xue L,Miller EL,Long DDE.Efficient metadata management in large distributed file systems.In:Miller E,Meter RV,eds.Proc.of the 20th IEEE/11th NASA Goddard Conf.on Mass Storage Systems and Technologies.San Diego:IEEE Computer Society,2003.290-298.
    [4]Weil SA,Pollack KT,Brandt SA,Miller EL.Dynamic metadata management for Petabyte-scale file systems.In:Benton V,ed.Proc.of the ACM/IEEE SC 2004 Conf.Washington:IEEE Computer Society,2004.4-15.
    [5]Liu Z,Zhou XM.A data object placement algorithm based on dynamic interval mapping.Journal of Software,2005,16(11):1886-1893 (in Chinese with English abstract).http://www.jos.org.cn/1000-9825/16/1886.htm
    [6]Liu Z,Zhou XM.An adaptive data objects placement algorithm for non-uniform capacities.In:Jin H,Pan Y,Xiao N,Sun JH,eds.Proc.of the 3rd Int'l Conf.on Grid and Cooperative Computing.LNCS 3251,Berlin,Heidelberg:Springer-Verlag,2004.423-430.
    [7]Liu Z,Xiao N,Zhou XM.RDIM:A self-adaptive and balanced distribution for replicated data in scalable storage clusters.In:Pan Y,Chen DX,Guo MY,Cao JN,Dongarra J,eds.Proc.of the 3rd Int'l Symp.on Parallel and Distributed Processing and Applications.LNCS 3758,Berlin,Heidelberg:Springer-Verlag,2005.21-32.
    [8]Ganger GR,Kaashoek MF.Embedded inodes and explicit groupings:Exploiting disk bandwidth for small files.In:Proc.of the 1997 USENIX Annual Technical Conf.Anaheim:USENIX,1997.1-17.
    [9]Litwin W,Neimat MA,Schneider DA.LH*-A scalable,distributed data structure.ACM Trans.on Database Systems,1996,21(4):480-525.
    [10]McKusick MK,Ganger GR.Soft updates:A technique for eliminating most synchronous writes in the fast file system.In:Proc.of the FREENIX Track:1999 USENIX Annual Technical Conf.Monterey:USENIX,1999.1-18.
    [11]Duchamp D.Optimistic lookup of whole NFS paths in a single operation.In:Proc.of the USENIX Summer 1994 Technical Conf.Boston:USENIX,1994.161-169.
    [5]刘仲,周兴铭.基于动态区间映射的数据对象布局算法.软件学报,2005,16(11):1886-1893.http://www.jos.org.cn/1000-9825/16/ 1886.htm
Get Citation

刘仲,周兴铭.基于目录路径的元数据管理方法.软件学报,2007,18(2):236-245

Copy
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:October 09,2005
  • Revised:December 31,2005
You are the firstVisitors
Copyright: Institute of Software, Chinese Academy of Sciences Beijing ICP No. 05046678-4
Address:4# South Fourth Street, Zhong Guan Cun, Beijing 100190,Postal Code:100190
Phone:010-62562563 Fax:010-62562533 Email:jos@iscas.ac.cn
Technical Support:Beijing Qinyun Technology Development Co., Ltd.

Beijing Public Network Security No. 11040202500063