Advertisement

Storage Virtualization System with Load Balancing for SAN

  • Weitao Sun
  • Ji-wu Shu
  • Weimin Zheng
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3252)

Abstract

Logical Volume Manager (LVM) has been a key subsystem for online disk storage management. Additional layer is created in the kernel to present a logical view of physical storage devices. Many transparent functions can be implemented between the logical and physical layers, such as merging several physical disks into a larger logical device, resizing logical devices without stopping the system. In a logical volume group, files can be striped into several physical disks so as to achieve high I/O performance. But data I/O parallelism by itself does not guarantee the optimal performance of an application since higher data throughput does not necessarily result in better application performance. This paper studied the dynamic load balancing and data redistribution algorithms in the storage virtualization layer when the load becomes imbalanced across the disks due to access pattern fluctuation. An extension of the heuristic load balancing method was proposed to the storage virtualization subsystem of Tsinghua-Mass Storage Network System (TH-MSNS). Logical volume I/O request status is monitored and the physical disks are sorted according to the access number of Logical Extents (LE) per time unit. The I/O operations on a LE of the hottest disk are transparently migrated to other disks. The preliminary performance simulations under a WWW server file access workload give satisfactory results by the promising cooling algorithm in storage virtualization systems.

Keywords

Access Pattern Cool Method Physical Volume Logical Volume Dynamic Load Balance 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Wilson, S.: Managing a Fibre Channel Storage Area Net work. Storage Networking Industry Association (November 1998)Google Scholar
  2. 2.
    Kim, M.Y.: Synchronized Disk Interleaving. IEEE trans. On computers C-35(11) (1986)Google Scholar
  3. 3.
    Salem, K., Garcia-Molina, H.: Disk Striping. In: Proc. Of the 2nd Int. Conf. On Data Engineering (1986)Google Scholar
  4. 4.
    Weikum, G., Zabback, P., Scheuermann, P.: Dynamic File Allocation in Disk Arrays. In: ACM SIGMOD Conf. (1991)Google Scholar
  5. 5.
    Tsai, H.P., Huang, J.L., Chao, C.M., et al.: SIFA: A Scalable File System with Intelligent File Allocation. In: Proc. of the 26th Int. COMPSAC (2002)Google Scholar
  6. 6.
    Scheuermann, P., Weikum, G., Zabback, P.: Data Partitioning and Load Balancing in Parallel Disk Systems. VLDB J. 7(1), 48–66 (1998)CrossRefGoogle Scholar
  7. 7.
    Awerbuch, B., Bartal, Y., Fiat, A.: Competitive Distributed File Allocation. Information and Computation 185, 1–40 (2003)zbMATHCrossRefMathSciNetGoogle Scholar
  8. 8.
    Ahmad, I., Karlapalem, K., Kwok, Y.K., et al.: Evolutionary Algorithms for Allocating Data in Distributed Database Systems. Distributed and Parallel Databases 11, 5–32 (2002)zbMATHCrossRefGoogle Scholar
  9. 9.
    Shu, J.W., et al.: A Highly Efficient FC-SAN Based on Load Stream. In: Zhou, X., Xu, M., Jähnichen, S., Cao, J. (eds.) APPT 2003. LNCS, vol. 2834, pp. 31–40. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  10. 10.
    Schwetman, H.: CSIM19: A Powerful Tool for Building System Models. In: Peters, B.A., Smith, J.S., Medeiros, D.J., Roher, M.W. (eds.) Proceedings of the 2001 Winter Simulation Conference, Washington, D.C., pp. 250–255 (2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Weitao Sun
    • 1
  • Ji-wu Shu
    • 1
  • Weimin Zheng
    • 1
  1. 1.Department of Computer Science and TechnologyTsinghua UniversityBeijingChina

Personalised recommendations