A New Data Sieving Approach for High Performance I/O

  • Yin Lu
  • Yong Chen
  • Prathamesh Amritkar
  • Rajeev Thakur
  • Yu Zhuang
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 164)

Abstract

Many scientific computing applications and engineering simulations exhibit noncontiguous I/O access patterns. Data sieving is an important technique to improve the performance of noncontiguous I/O accesses by combining small and noncontiguous requests into a large and contiguous request. It has been proven effective even though more data is potentially accessed than demanded. In this study, we propose a new data sieving approach namely Performance Model Directed Data Sieving, or PMD data sieving in short. It improves the existing data sieving approach from two aspects: (1) dynamically determines when it is beneficial to perform data sieving; and (2) dynamically determines how to perform data sieving if beneficial. It improves the performance of the existing data sieving approach and reduces the memory consumption as verified by experimental results. Given the importance of supporting noncontiguous accesses effectively and reducing the memory pressure in a large-scale system, the proposed PMD data sieving approach in this research holds a promise and will have an impact on high performance I/O systems.

Keywords

Data sieving Runtime systems Parallel I/O Libraries Parallel file systems High performance computing 

References

  1. 1.
    Blas, J.G., Isaila, F., Carretero, J., Latham, R., Ross. R.: Multiple-level MPI file write-back and prefetching for blue gene systems. In: Proceedings of the PVM/MPI (2009)Google Scholar
  2. 2.
    Bordawekar, R., Rosario, J.M., Choudhary, A.N.: Design and evaluation of primitives for parallel I/O. In: Proceedings of the ACM/IEEE Supercomputing Conference (1993)Google Scholar
  3. 3.
    Carns, P.H., Ligon, W.B., III, Ross, R.B., Thakur, R.: PVFS “a parallel file system for linux clusters.” In: Proceedings of the 4th Annual Linux Showcase and Conference (2000)Google Scholar
  4. 4.
    Cluster File Systems Inc.: Lustre: a scalable, high performance file system. Whitepaper. http://www.lustre.org/docs/whitepaper.pdf
  5. 5.
    Crandall, P.E., Aydt, R.A., Chien, A.A., Reed, D.A.: Input/output characteristics of scalable parallel applications. In: Proceedings of the ACM/IEEE Conference on Supercomputing, pp. 59-es (1995)Google Scholar
  6. 6.
    Iskra, K., Romein, J.W., Yoshii, K., Beckman, P.: ZOID: I/O forwarding infrastructure for petascale architectures. In: Proceedings of the 13th ACM PPoPP (2008)Google Scholar
  7. 7.
    Lei, H., Duchamp, D.: An analytical approach to file prefetching. In: Proceedings of the 1997 USENIX Annual Technical Conference, pp. 275–288, Jan 1997Google Scholar
  8. 8.
    Lofstead, J.F., Klasky, S., Schwan, K., Podhorszki, N., Jin, C.: Flexible I/O and integration for scientific codes through the adaptable I/O system (ADIOS). In: Proceedings of the 6th International Workshop on Challenges of Large Applications in Distributed Environments (2008)Google Scholar
  9. 9.
    May, J.: Parallel I/O for high performance computing. Morgan Kaufmann, San Francisco (2001)Google Scholar
  10. 10.
    Ma, X.S., Winslett, M., et. al.: Faster collective output through active buffering. In: IPDPS (2002)Google Scholar
  11. 11.
    Nisar,A., Liao, W.-K., Choudhary, A.: Scaling parallel I/O performance through I/O delegate and caching system. SC (2008)Google Scholar
  12. 12.
    Nitzberg, B., et al.: Collective buffering: improving parallel I/O performance. In: HPDC (1997)Google Scholar
  13. 13.
    Rafique, M.M., Butt, A.R., Nikolopoulos, D.S.: DMA-based prefetching for I/O-intensive workloads on the cell architecture. In: Conference on Computing Frontiers, pp. 23–32 (2008)Google Scholar
  14. 14.
  15. 15.
    Schmuck, F., Haskin, R., GPFS: a shared-disk file system for large computing clusters. In: Proceedings of the First USENIX FAST, pp. 231–244, USENIX, Jan 2002Google Scholar
  16. 16.
    Tran, N., Reed, D.A.: Automatic ARIMA time series modeling for adaptive I/O prefetching. IEEE Trans. Parallel Distrib. Sys. 15(4), 362–377 (2004)CrossRefGoogle Scholar
  17. 17.
    Thakur, R., Gropp, W., Lusk, E.: Data sieving and collective I/O in ROMIO. In: Proceedings of the 7th Symposium on the Frontiers of Massively Parallel Computation (1999)Google Scholar
  18. 18.
    Thakur, R., Choudhary, A., Bordawekar, R., More, S., Kuditipudi, S.: Passion: optimized I/O for parallel applications. Computer 29(6), 70–78, June 1996Google Scholar
  19. 19.
    Vilayannur, M., Sivasubramaniam, A., Kandemir, M.T., Thakur, R., Ross, R.: Discretionary caching for I/O on clusters. Cluster Comput. 9(1), 29–44 (2006)CrossRefGoogle Scholar
  20. 20.
    Welch, B., Unangst, M., Abbasi, Z., Gibson, G., Mueller, B., Small, J., Zelenka, J., Zhou, B.: Scalable performance of the panasas parallel file system. USENIX FAST (2008)Google Scholar
  21. 21.
    Zhang, X., Jiang, S., Davis, K.: Making resonance a common case: a high-performance implementation of collective I/O on parallel file systems. IPDPS (2009)Google Scholar

Copyright information

© Springer Science+Business Media Dortdrecht 2012

Authors and Affiliations

  • Yin Lu
    • 1
  • Yong Chen
    • 1
  • Prathamesh Amritkar
    • 1
  • Rajeev Thakur
    • 2
  • Yu Zhuang
    • 1
  1. 1.Computer Science DepartmentTexas Tech UniversityTXUSA
  2. 2.Mathematics and Computer Science DivisionArgonne National LabArgonneUSA

Personalised recommendations