Dynamic Stripe Management Mechanism in Distributed File Systems

  • Jianwei Liao
  • Guoqiang Xiao
  • Xiaoyan Liu
  • Lingyu Zhu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8707)


This paper presents a novel mechanism to dynamically re-size and re-distribute stripes on the storage servers in distributed file systems. To put this mechanism to work, the information about logical I/O access on the client side is piggybacked to physical I/O access on the storage server side, for building the relationship between the logical I/O access and physical I/O access. Moreover, this newly presented mechanism supports varying size of stripes on the storage servers to obtain finer concurrency granularity on accessing to data stripes. As a result, the mapping relationship can be utilized to direct stripe re-sizing and re-distributing on the storage servers dynamically for better system performance. Experimental results show that this stripe management mechanism can reduce I/O response time and boost I/O data throughput significantly for applications with complicated access patterns.


Distributed/parallel file systems Re-sizing and re- distributing stripes Varying stripe size I/O optimization 


  1. 1.
    Gantz, J., Reinsel, D.: The digital universe in 2020: Big Data, Bigger Digital Shadows, Biggest Growth in the Far East, United States (2013), (accessed on October 3, 2013)
  2. 2.
    Digital database for screening mammography, (accessed on December 12, 2011)
  3. 3.
    MADbench2. borrill/MADbench2/,
  4. 4.
    Weil, S.A., Pollack, K.T., Brandt, S.A., Miller, E.L.: Dynamic metadata management for petabyte-scale file systems. In: Proceedings of the 2004 ACM/IEEE Conference on Supercomputing, SC 2014, pp. 4–15. IEEE Computer Society, Washington, DC (2004)Google Scholar
  5. 5.
    Nieuwejaar, N., Kotz, D.: The galley parallel file system. Parallel Computing 23(4-5), 447–476 (1997)CrossRefzbMATHGoogle Scholar
  6. 6.
    Kunkel, J., Ludwig, T.: Performance evaluation of the pvfs2 architecture. In: Proceedings of 15th EUROMICRO International Conference on Parallel, Distributed and Network-Based Processing (PDP 200), pp. 509–516 (2007)Google Scholar
  7. 7.
    Liao, J., Ishikawa, Y.: Partial replication of metadata to achieve high metadata availability in parallel file systems. In: Proceedings of the 41st International Conference on Parallel Processing, ICPP 2012, pp. 168–177 (2012)Google Scholar
  8. 8.
    Latham, R., Miller, N., Ross, R., Carns, P.: A Next- Generation Parallel File System for Linux Clusters. Linux World 2(1) (2004)Google Scholar
  9. 9.
    Schmuck, F., Haskin, R.: Gpfs: A shared-disk file system for large computing clusters. In: Proceedings of the 1st USENIX Conference on File and Storage Technologies, FAST 2002. USENIX Association, Berkeley (2002)Google Scholar
  10. 10.
    Schwan, P.: Lustre: Building a file system for 1,000-node clusters. In: Proceedings of the Linux Symposium, p. 9 (2003)Google Scholar
  11. 11.
    Ghemawat, S., Gobioff, H., Leung, T.: The Google file system. ACM SIGOPS Operating Systems Review 37(5), 29–43 (2003)CrossRefGoogle Scholar
  12. 12.
    Liao, J.: Self-tuning optimization on storage servers in parallel file system. Journal of Circuits, Systems and Computers 30(4), 21 pages (2014)Google Scholar
  13. 13.
    Medina, M.: A self-tuning disk striping system for parallel input/output. Dissertation. University of Illinois at Urbana-Champaign, USA (2007)Google Scholar
  14. 14.
    Byna, S., Chen, Y., Sun, X.-H., Thakur, R., Gropp, W.: Parallel i/o prefetching using mpi file caching and i/o signatures. In: SC 2008, pp. 44:1-44:12 (2008)Google Scholar
  15. 15.
    Madhyastha, T.: Automatic Classification of Input/Output Acess Patterns. Dissertation, Champaign, IL, USA (1997)Google Scholar
  16. 16.
    Li, Z., Chen, Z., Srinivasan, S., Zhou, Y.: C-Miner: Mining Block Correlations in Storage Systems. In: Proceedings of the 3rd Conference on File and Storage Technologies, FAST 2004 (2004)Google Scholar
  17. 17.
    Li, Z., Chen, Z., Zhou, Y.: Mining Block Correlations to Improve Storage Performance. ACM Transactions on Storage 1(1), 213–245 (2005)CrossRefGoogle Scholar
  18. 18.
    Narayan, S., Chandy, J.: Trace Based Analysis of File System Effects on Disk I/O. In: Proceedings of 2004 International Symposium on Performance Evaluation of Computer and Telecommunication Systems, SPECTS 2004 (2004)Google Scholar
  19. 19.
    Narayan, S.: File System Optimization Using Block Reorganization Techniques. Master of Science Thesis, University of Connecticut (2004)Google Scholar
  20. 20.
    Hsu, W., Smith, A., Young, H.: The automatic improvement of locality in storage systems. ACM Trans. Comput. Syst. 23(4), 424–473 (2005)CrossRefGoogle Scholar
  21. 21.
    Jiang, S., Ding, X., Xu, Y., Davis, K.: A Prefetching Scheme Exploiting both Data Layout and Access History on Disk. ACM Transaction on Storage 9(3), Article 10, 23 p. (2013)Google Scholar
  22. 22.
    Song, H., Yin, Y., Sun, X., Thakur, R., Lang, S.: Server-side I/O coordination for parallel file systems. In: Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC 2011). ACM (2011)Google Scholar
  23. 23.
    He, J., Bent, J., Torres, A., Sun, X., et al.: I/O Acceleration with Pattern Detection. In: Proceedings of the 22nd International ACM Symposium on High Performance Parallel and Distributed Computing (HPDC 2013), pp. 26-35 (2013)Google Scholar
  24. 24.
    Simitci, H.: Adaptive Disk Striping for Parallel Input/Output. Dissertation, Champaign (2000)Google Scholar
  25. 25.
    Dong, B., Li, X., Xiao, L., et al.: A New File-Specific Stripe Size Selection Method for Highly Concurrent Data Access. In: Proceedings of 2012 ACM/IEEE 13th International Conference on Grid Computing (GRID), pp. 22-30 (2012)Google Scholar
  26. 26.
    Triantafillou, P., Faloutsos, C.: Overlay striping and optimal parallel I/O for modern applications. Parallel Computing 24(1), 21–43 (1998) Special Issue on Applications: Parallel Data Servers and Applications (1998)Google Scholar

Copyright information

© IFIP International Federation for Information Processing 2014

Authors and Affiliations

  • Jianwei Liao
    • 1
    • 2
  • Guoqiang Xiao
    • 1
  • Xiaoyan Liu
    • 1
  • Lingyu Zhu
    • 1
  1. 1.College of Computer and Information ScienceSouthwest University of ChinaBeibeiP.R. China
  2. 2.State Key Laboratory for Novel Software TechnologyNanjing UniversityNanjingP.R. China

Personalised recommendations