Advertisement

Analysis of Grid Storage Element Architectures: High-end Fiber-Channel vs. Emerging Cluster-based Networked Storage

  • Maciej Brzezniak
  • Norbert Meyer
  • Michail D. Flouris
  • Renaud Lachaiz
  • Angelos Bilas

Storage elements that can scale to large capacities and high-performance are an essential component of future GRID infrastructures, especially for supporting an increasing number of data-intensive applications and services. This paper studies two approaches for building scalable networked storage elements: enterpriselevel, Fibre-Channel-based Storage (FCS) and commodity, Cluster-based Networked Storage (CNS). First we review the characteristics of FCS, which is currently widely used in high-end enterprise-level installations, discussing various aspects, such as scalability, performance, availability, manageability and security. Then, we compare it with CNS and consider how features of high-end specialized systems may be provided on top of this new architecture. We believe that CNS has a potential for replacing FCS in many application domains; however, there is a need for addressing the feature gap between FCS and CNS at the architectural and storage management layer.

Keywords

Storage systems storage management Storage Element 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

 References

  1. [1]
    Storage Magazine, Alan Radding: Is encryption enough?, http://storagemagazine.techtarget.com/magItem/0,291266,sid35 gci1193200,00.html.
  2. [2]
    SNIA, A SNIA Dictionary of Storage Networking Terminology., http://www.snia.org/education/dictionary .
  3. [3]
    SNIA, The SNIA’s Storage Management Initiative., http://www.snia.org/smi/home. [4] Top500 supercomputing sites.
  4. [5]
    Woven Systems,The Woven Systems EFX1000Ethernet Fabric Switch., http://wovensystems.com.
  5. [6]
    Xyratex Technology Limited., RS-4835-E3-EBD Datasheet., http://www.xyratex.com/ products/storage-systems/storage-RS-4835-E3-EBD.asp.
  6. [7]
    H. bung Chen, G. Grider, and P. Fields, A Cost-Effective, High Bandwidth Server I/O network Architecture for Cluster Systems., in International Parallel and Distributed Processing Symposium (IPDPS 2007), pp. 1-10, IEEE, 2007.Google Scholar
  7. [8]
    B. Schroeder and C. M. U. Garth A. Gibson, Disk Failures in the Real World: What Does an MTTF of 1,000,000 Hours Mean to You?, in Proceedings of 5th USENIX Conference on File and Storage Technologies (FAST ’07), pp. 1-16, San Francisco, CA, USA, 2007.Google Scholar
  8. [9]
    B. Schroeder and C. M. U. Garth A. Gibson, Disk Failures in the Real World: What Does an MTTF of 1,000,000 Hours Mean to You?, in Proceedings of 5th USENIX Conference on File and Storage Technologies (FAST ’07), pp. 17-28, San Francisco, CA, USA, 2007.Google Scholar
  9. [10]
    Y. Saito, S. Frolund, A. Veitch, A. Merchant, and S. Spence, FAB: Enterprise storage systems on a shoestring, in Proc. of the ASPLOS 2004, 2004.Google Scholar
  10. [11]
    M. W. Storer, K. Greenan, E. L. Miller, and K. Voruganti, POTSHARDS: Secure Long- Term Storage Without Encryption, in 2007 USENIX Annual Technical Conference, 2007.Google Scholar
  11. [12]
    A. J. K. Gregory R. Ganger, John D. Strunk, Carnegie Mellon University Report No. CMU-CS-03-178, 2003 (unpublished).Google Scholar
  12. [13]
    N. P. Kronenberg, H. M. Levy, and W. D. Strecker, ACM Transactions on Computer Systems 4 (1986).Google Scholar
  13. [14]
    S. Soltis, G. Erickson, K. Preslan, M. O’Keefe, and T. Ruwart, The Global File System: A File System for Shared Disk Storage, 1997.Google Scholar
  14. [15]
    F. Schmuck and R. Haskin, GPFS: A Shared-disk File System for Large Computing Centers, in USENIX Conference on File and Storage Technologies, pp. 231-244, Monterey, CA, 2002.Google Scholar
  15. [16]
    M. A.-E.-M. et al., Ursa Minor: Versatile Cluster-Based Storage, in Proceedings of the 4th USENIX Conference on File and Storage Technology, San Francisco, CA, USA, 2005.Google Scholar
  16. [17]
    K. A. Amiri, G. A. Gibson, and R. Golding, Highly Concurent Shared Storage, in Proceed- ings of 20th International Conference on Distributed Computing Systems (ICDCS’2000), edited by IEEE, pp. 298-307, Taipe, Taiwan, R.O.C, 2000, IEEE Computer.Google Scholar
  17. [18]
    C. R. Lumb, R. Golding, and G. R. Ganger, D-SPTF: Decentralized Request Distribution in Brick-Based Storage Systems, in Proceedings of the 11th ACM ASPLOS Conference, Boston, MA, USA, 2004.Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  • Maciej Brzezniak
    • 1
  • Norbert Meyer
    • 1
  • Michail D. Flouris
    • 2
  • Renaud Lachaiz
    • 2
  • Angelos Bilas
    • 2
  1. 1.Poznan Supercomputing and Networking CenterPoland
  2. 2.Institute of Computer ScienceFoundation for Research and Technology – HellasGreece

Personalised recommendations