Improving Read Performance with Online Access Pattern Analysis and Prefetching

  • Houjun Tang
  • Xiaocheng Zou
  • John Jenkins
  • David A. BoyukaII
  • Stephen Ranshous
  • Dries Kimpe
  • Scott Klasky
  • Nagiza F. Samatova
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8632)

Abstract

Among the major challenges of transitioning to exascale in HPC is the ubiquitous I/O bottleneck. For analysis and visualization applications in particular, this bottleneck is exacerbated by the write-onceread- many property of most scientific datasets combined with typically complex access patterns. One promising way to alleviate this problem is to recognize the application’s access patterns and utilize them to prefetch data, thereby overlapping computation and I/O. However, current research methods for analyzing access patterns are either offline-only and/or lack the support for complex access patterns, such as high-dimensional strided or composition-based unstructured access patterns. Therefore, we propose an online analyzer capable of detecting both simple and complex access patterns with low computational and memory overhead and high accuracy. By combining our pattern detection with prefetching,we consistently observe run-time reductions, up to 26%, across 18 configurations of PIOBench and 4 configurations of a micro-benchmark with both structured and unstructured access patterns.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Chen, J.H., Choudhary, A., De Supinski, B., DeVries, M., Hawkes, E., Klasky, S., Liao, W., Ma, K., Mellor-Crummey, J., Podhorszki, N., et al.: Terascale direct numerical simulations of turbulent combustion using S3D. Computational Science & Discovery 2(1), 15001 (2009)CrossRefGoogle Scholar
  2. 2.
    Wang, W., Lin, Z., Tang, W., Lee, W., Ethier, S., Lewandowski, J., Rewoldt, G., Hahm, T., Manickam, J.: Gyro-kinetic simulation of global turbulent transport properties in tokamak experiments. Physics of Plasmas 13, 092505 (2006)Google Scholar
  3. 3.
    Zhu, Y., Jiang, H., Qin, X., Feng, D., Swanson, D.R.: Improved read performance in a cost-effective, fault-tolerant parallel virtual file system (ceft-pvfs). In: CCGrid 2003, pp. 730–735. IEEE (2003)Google Scholar
  4. 4.
    Di Biagio, A., Speziale, E., Agosta, G.: Exploiting thread-data affinity in openmp with data access patterns. In: Jeannot, E., Namyst, R., Roman, J. (eds.) Euro-Par 2011, Part I. LNCS, vol. 6852, pp. 230–241. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  5. 5.
    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. 1–12. IEEE (2008)Google Scholar
  6. 6.
    Oly, J., Reed, D.A.: Markov model prediction of I/O requests for scientific applications. In: ICS 2002, pp. 147–155. ACM (2002)Google Scholar
  7. 7.
    Li, Z., Chen, Z., Srinivasan, S.M., Zhou, Y.: C-Miner: Mining Block Correlations in Storage Systems. In: FAST, pp. 173–186 (2004)Google Scholar
  8. 8.
    Choi, J.Y., Abbasi, H., Pugmire, D., Podhorszki, N., Klasky, S., Capdevila, C., Parashar, M., Wolf, M., Qiu, J., Fox, G.: Mining hidden mixture context with adios-p to improve predictive pre-fetcher accuracy. In: 2012 IEEE 8th International Conference on E-Science (e-Science), pp. 1–8. IEEE (2012)Google Scholar
  9. 9.
    Crandall, P.E., Aydt, R.A., Chien, A.A., Reed, D.A.: Input/output characteristics of scalable parallel applications. In: Proceedings of the IEEE/ACM SC 1995 Conference on Supercomputing, pp. 59–59. IEEE (1995)Google Scholar
  10. 10.
    Madhyastha, T.M., Reed, D.A.: Learning to classify parallel input/output access patterns. TPDS 13(8), 802–813 (2002)Google Scholar
  11. 11.
    Carns, P., Latham, R., Ross, R., Iskra, K., Lang, S., Riley, K.: 24/7 characterization of petascale I/O workloads. In: Cluster 2010, pp. 1–10 (2010)Google Scholar
  12. 12.
    Shorter, F.: Design and analysis of a performance evaluation standard for parallel file systems. PhD thesis, Clemson University (2003)Google Scholar
  13. 13.
    Gong, Z., Boyuka, D., Zou, X., Liu, Q., Podhorszki, N., Klasky, S., Ma, X., Samatova, N.F.: Parlo: Parallel run-time layout optimization for scientific data explorations with heterogeneous access patterns. In: CCGrid 2013, pp. 343–351 (2013)Google Scholar
  14. 14.
    Han, W.S., Moon, Y.S., Whang, K.Y.: Prefetchguide: Capturing navigational access patterns for prefetching in client/server object-oriented/object-relational dbmss. Information Sciences 152, 47–61 (2003)CrossRefGoogle Scholar
  15. 15.
    Baer, J.L., Chen, T.F.: An effective on-chip preloading scheme to reduce data access penalty. In: Proceedings of the 1991 ACM/IEEE Conference on Supercomputing 1991, pp. 176–186. IEEE (1991)Google Scholar
  16. 16.
    Dahlgren, F., Dubois, M., Stenstrom, P.: Fixed and adaptive sequential prefetching in shared memory multiprocessors. In: ICPP 1993, vol. 1, pp. 56–63. IEEE (1993)Google Scholar
  17. 17.
    Dahlgren, F., Dubois, M., Stenstrom, P.: Sequential hardware prefetching in shared-memory multiprocessors. TPDS 6(7), 733–746 (1995)Google Scholar
  18. 18.
    Ding, X., Jiang, S., Chen, F., Davis, K., Zhang, X.: Diskseen: Exploiting disk layout and access history to enhance I/O prefetch. In: USENIX Annual Technical Conference, vol. 7, pp. 261–274 (2007)Google Scholar
  19. 19.
    Carns, P.H., Ligon III, W.B., Ross, R.B., Thakur, R.: Pvfs: A parallel file system for linux clusters. In: Proceedings of the 4th Annual Linux Showcase and Conference, pp. 391–430 (2000)Google Scholar
  20. 20.
    Braam, P.J., Zahir, R.: Lustre: A scalable, high performance file system. Cluster File Systems, Inc. (2002)Google Scholar
  21. 21.
    Patterson, R.H., Gibson, G.A., Ginting, E., Stodolsky, D., Zelenka, J.: Informed prefetching and caching, vol. 29. ACM (1995)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Houjun Tang
    • 1
    • 2
  • Xiaocheng Zou
    • 1
    • 2
  • John Jenkins
    • 1
    • 3
  • David A. BoyukaII
    • 1
    • 2
  • Stephen Ranshous
    • 1
    • 2
  • Dries Kimpe
    • 3
  • Scott Klasky
    • 2
  • Nagiza F. Samatova
    • 1
    • 2
  1. 1.North Carolina State UniversityRaleighUSA
  2. 2.Oak Ridge National LaboratoryOak RidgeUSA
  3. 3.Argonne National LaboratoryArgonneUSA

Personalised recommendations