Data Migration in RAID Based on Stripe Unit Heat

  • Yan Liu
  • Chang-Sheng Xie
  • Huai-yang Li
  • Zhen Zhao
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3930)


Traditional RAID has the characteristics that location of stripe unit in each disk is stochastic and static, and that the outer zone of the disk has higher data transfer rate as compared to the inner one. Facing this situation, to exploit RAID I/O performance fully, this paper proposes a new algorithm PMSH (Placement and Migration based on Stripe unit Heat) for RAID stripe unit data to be placed optically and migrated dynamically. Based on the heat of RAID stripe unit, PMSH keeps migrating the frequently accessed stripe unit to the disk zone with higher data transfer rate to optimize the location of data in RAID disks and make the data distribution adapt to the evolution of file access pattern dynamically as well. Simulation results demonstrate significant RAID I/O performance improvement using PMSH.


Outer Zone Data Migration Access Rate Data Layout Data Transfer Rate 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Rummer, C., Wildes, J.: Introduction to disk drive modeling. IEEE Computer 27, 17–29 (1993)Google Scholar
  2. 2.
    Meter, R.V.: Observing the effects of multi-zone disks. In: USENIX (ed.) Annual Technical Conference, Anaheim, CA, pp. 19–30. USENIX, Berkeley (1997)Google Scholar
  3. 3.
    Seagate. Seagate Web Page (2000),
  4. 4.
    Patterson, D.A., Chen, P., Gibson, G., Katz, R.H.: Introduction to Redundant Array of Independent Disk. In: 34th IEEE Computer Society International Conference, Intellectual Leverage, Digest of Papers, pp. 112–117 (1989)Google Scholar
  5. 5.
    Sivan - Zimet, M.: A comparison of access pattern (92-99). UCSC CS290S Project (2000)Google Scholar
  6. 6.
    Ghandeharizadeh, S., Ierardi, D., Kim, D., Zimmermann, R.: Placement of Data in Multi-Zone Disk Drives. In: Second International Baltic Workshop on DB and IS (1996)Google Scholar
  7. 7.
    Wang, J., Hu, Y.: PROFS-Performance-Oriented Data Reorganization for Log-Structured File System on Multi-Zone. In: Proceedings of Ninth International Symposium on Modeling Analysis and Simulation of Computer and Telecommunication Systems, pp. 285–292 (2001)Google Scholar
  8. 8.
    Chen, S., Thapar, M.: Zone-bit-recording-enhanced video data layout strategies. In: Proceedings of the 4th Int’l Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (1996)Google Scholar
  9. 9.
    Nerjes, G., Muth, P., Weikum, G.: Stochastic service guarnantees for continuous data on multi-zone disks. In: Proceedings of the Sixteenth ACM SIGACT-SIGMOD-SIGART Symposim on Principles of Database Systems, Tucson, Arizona, pp. 154–160 (1997)Google Scholar
  10. 10.
    Ganger, G.R., Patt, Y.N.: Using system-level models to evaluate I/O subsystems designs. IEEE Transactions on Computers, 667–678 (1998)Google Scholar
  11. 11.
    Gabber, E., Korth, H.F.: Data Logging: A Method for Efficient Data Updates in Constantly Active RAIDs. In: Proceedings of 14th International Conference on Data Engineering, pp. 144–153 (1998)Google Scholar
  12. 12.
    Menon, J.: A Performance Comparison of RAID-5 and Log-structured Arrays. In: Proceedings of the 4th IEEE International Symposium, pp. 167–178 (1995)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Yan Liu
    • 1
  • Chang-Sheng Xie
    • 1
  • Huai-yang Li
    • 1
  • Zhen Zhao
    • 1
  1. 1.Key laboratory of Data Storage System, School of Computer Science and TechnologyHuazhong University of Science and TechnologyWuhanChina

Personalised recommendations