Skip to main content

Storage wall for exascale supercomputing

Abstract

The mismatch between compute performance and I/O performance has long been a stumbling block as supercomputers evolve from petaflops to exaflops. Currently, many parallel applications are I/O intensive, and their overall running times are typically limited by I/O performance. To quantify the I/O performance bottleneck and highlight the significance of achieving scalable performance in peta/exascale supercomputing, in this paper, we introduce for the first time a formal definition of the ‘storage wall’ from the perspective of parallel application scalability. We quantify the effects of the storage bottleneck by providing a storage-bounded speedup, defining the storage wall quantitatively, presenting existence theorems for the storage wall, and classifying the system architectures depending on I/O performance variation. We analyze and extrapolate the existence of the storage wall by experiments on Tianhe-1A and case studies on Jaguar. These results provide insights on how to alleviate the storage wall bottleneck in system design and achieve hardware/software optimizations in peta/exascale supercomputing.

This is a preview of subscription content, access via your institution.

References

  • Agarwal, S., Garg, R., Gupta, M.S., et al., 2004. Adaptive incremental checkpointing for massively parallel systems. Proc. 18th Annual Int. Conf. on Supercomputing, p.277–286. http://dx.doi.org/10.1145/1006209.1006248

    Google Scholar 

  • Agerwala, T., 2010. Exascale computing: the challenges and opportunities in the next decade. IEEE 16th Int. Symp. on High Performance Computer Architecture. http://dx.doi.org/10.1109/HPCA.2010.5416662

    Google Scholar 

  • Alam, S.R., Kuehn, J.A., Barrett, R.F., et al., 2007. Cray XT4: an early evaluation for petascale scientific simulation. Proc. ACM/IEEE Conf. on Supercomputing, p.1–12. http://dx.doi.org/10.1145/1362622.1362675

    Google Scholar 

  • Ali, N., Carns, P.H., Iskra, K., et al., 2009. Scalable I/O forwarding framework for high-performance computing systems. IEEE Int. Conf. on Cluster Computing and Workshops, p.1–10, http://dx.doi.org/10.1109/CLUSTR.2009.5289188

    Google Scholar 

  • Amdahl, G.M., 1967. Validity of the single processor approach to achieving large scale computing capabilities. Proc. Spring Joint Computer Conf., p.483–485. http://dx.doi.org/10.1145/1465482.1465560

    Google Scholar 

  • Bent, J., Gibson, G., Grider, G., et al., 2009. PLFS: a checkpoint file system for parallel applications. Proc. Conf. on High Performance Computing Networking, Storage and Analysis, p.21. http://dx.doi.org/10.1145/1654059.1654081

    Google Scholar 

  • Cappello, F., Geist, A., Gropp, B., et al., 2009. Toward exascale resilience. Int. J. High Perform. Comput. Appl., 23(4):374–388. http://dx.doi.org/10.1177/1094342009347767

    Article  Google Scholar 

  • Carns, P., Harms, K., Allcock, W., et al., 2011. Understanding and improving computational science storage access through continuous characterization. ACM Trans. Stor., 7(3):1–26. http://dx.doi.org/10.1145/2027066.2027068

    Article  Google Scholar 

  • Chen, J., Tang, Y.H., Dong, Y., et al., 2016. Reducing static energy in supercomputer interconnection networks using topology-aware partitioning. IEEE Trans. Comput., 65(8):2588–2602. http://dx.doi.org/10.1109/TC.2015.2493523

    Article  MathSciNet  Google Scholar 

  • Culler, D.E., Singh, J.P., Gupta, A., 1998. Parallel Computer Architecture: a Hardware/Software Approach. Morgan Kaufmann Publishers Inc., San Francisco, USA.

    Google Scholar 

  • Egwutuoha, I.P., Levy, D., Selic, B., et al., 2013. A survey of fault tolerance mechanisms and checkpoint/restart implementations for high performance computing systems. J. Supercomput., 65(3):1302–1326. http://dx.doi.org/10.1007/s11227-013-0884-0

    Article  Google Scholar 

  • Elnozahy, E.N., Plank, J.S., 2004. Checkpointing for peta-scale systems: a look into the future of practical rollback-recovery. IEEE Trans. Depend. Secur. Comput., 1(2):97–108. http://dx.doi.org/10.1109/TDSC.2004.15

    Article  Google Scholar 

  • Elnozahy, E.N., Alvisi, L., Wang, Y.M., et al., 2002. A survey of rollback-recovery protocols in message-passing systems. ACM Comput. Surv., 34(3):375–408. http://dx.doi.org/10.1145/568522.568525

    Article  Google Scholar 

  • Fahey, M., Larkin, J., Adams, J., 2008. I/O performance on a massively parallel cray XT3/XT4. IEEE Int. Symp. on Parallel and Distributed Processing, p.1–12. http://dx.doi.org/10.1109/IPDPS.2008.4536270

    Google Scholar 

  • Ferreira, K.B., Riesen, R., Bridges, P., et al., 2014. Accelerating incremental checkpointing for extreme-scale computing. Fut. Gener. Comput. Syst., 30:66–77. http://dx.doi.org/10.1016/j.future.2013.04.017

    Article  Google Scholar 

  • Frasca, M., Prabhakar, R., Raghavan, P., et al., 2011. Virtual I/O caching: dynamic storage cache management for concurrent workloads. Proc. Int. Conf. for High Performance Computing, Networking, Storage and Analysis, p.38. http://dx.doi.org/10.1145/2063384.2063435

    Google Scholar 

  • Gamblin, T., de Supinski, B.R., Schulz, M., et al., 2008. Scalable load-balance measurement for SPMD codes. Proc. ACM/IEEE Conf. on Supercomputing, p.1–12.

    Google Scholar 

  • Gustafson, J.L., 1988. Reevaluating Amdahl’s law. Commun. ACM, 31(5):532–533. http://dx.doi.org/10.1145/42411.42415

    Article  Google Scholar 

  • Hargrove, P.H., Duell, J.C., 2006. Berkeley lab checkpoint/restart (BLCR) for Linux clusters. J. Phys. Conf. Ser., 46(1):494–499. http://dx.doi.org/10.1088/1742-6596/46/1/067

    Article  Google Scholar 

  • Hennessy, J.L., Patterson, D.A., 2011. Computer Architecture: a Quantitative Approach. Elsevier.

    MATH  Google Scholar 

  • HPCwire, 2010. DARPA Sets Ubiquitous HPC Program in Motion. Available from http://www.hpcwire.com/2010/08/10/darpa_sets_ubiquitous_hpc_program_ in_motion/.

    Google Scholar 

  • Hu, W., Liu, G.M., Li, Q., et al., 2016. Storage speedup: an effective metric for I/O-intensive parallel application. 18th Int. Conf. on Advanced Communication Technology, p.1–2. http://dx.doi.org/10.1109/ICACT.2016.7423395

    Google Scholar 

  • Kalaiselvi, S., Rajaraman, V., 2000. A survey of checkpointing algorithms for parallel and distributed computers. Sadhana, 25(5):489–510. http://dx.doi.org/10.1007/BF02703630

    Article  Google Scholar 

  • Kim, Y., Gunasekaran, R., 2015. Understanding I/O workload characteristics of a peta-scale storage system. J. Supercomput., 71(3):761–780. http://dx.doi.org/10.1007/s11227-014-1321-8

    Article  Google Scholar 

  • Kim, Y., Gunasekaran, R., Shipman, G.M., et al., 2010. Workload characterization of a leadership class storage cluster. Petascale Data Storage Workshop, p.1–5. http://dx.doi.org/10.1109/PDSW.2010.5668066

    Google Scholar 

  • Kotz, D., Nieuwejaar, N., 1994. Dynamic file-access characteristics of a production parallel scientific workload. Proc. Supercomputing, p.640–649. http://dx.doi.org/10.1109/SUPERC.1994.344328

    Chapter  Google Scholar 

  • Liao, W.K., Ching, A., Coloma, K., et al., 2007. Using MPI file caching to improve parallel write performance for large-scale scientific applications. Proc. ACM/IEEE Conf. on Supercomputing, p.8. http://dx.doi.org/10.1145/1362622.1362634

    Google Scholar 

  • Liu, N., Cope, J., Carns, P., et al., 2012. On the role of burst buffers in leadership-class storage systems. IEEE 28th Symp. on Mass Storage Systems and Technologies, p.1–11. http://dx.doi.org/10.1109/MSST.2012.6232369

    Google Scholar 

  • Liu, Y., Gunasekaran, R., Ma, X.S., et al., 2014. Automatic identification of application I/O signatures from noisy server-side traces. Proc. 12th USENIX Conf. on File and Storage Technologies, p.213–228.

    Google Scholar 

  • Lu, K., 1999. Research on Parallel File Systems Technology Toward Parallel Computing. PhD Thesis, National University of Defense Technology, Changsha, China (in Chinese).

    Google Scholar 

  • Lucas, R., Ang, J., Bergman, K., et al., 2014. DOE Advanced Scientific Computing Advisory Subcommittee (ASCAC) Report: Top Ten Exascale Research Challenges. USDOE Office of Science. http://dx.doi.org/10.2172/1222713

    Google Scholar 

  • Miller, E.L., Katz, R.H., 1991. Input/output behavior of supercomputing applications. Proc. ACM/IEEE Conf. on Supercomputing, p.567–576. http://dx.doi.org/10.1145/125826.126133

    Google Scholar 

  • Moreira, J., Brutman, M., Castano, J., et al., 2006. Designing a highly-scalable operating system: the blue Gene/L story. Proc. ACM/IEEE Conf. on Supercomputing, p.53–61. http://dx.doi.org/10.1109/SC.2006.23

    Google Scholar 

  • Oldfield, R.A., Arunagiri, S., Teller, P.J., et al., 2007. Modeling the impact of checkpoints on next-generation systems. 24th IEEE Conf. on Mass Storage Systems and Technologies, p.30–46. http://dx.doi.org/10.1109/MSST.2007.4367962

    Google Scholar 

  • Pasquale, B.K., Polyzos, G.C., 1993. A static analysis of I/O characteristics of scientific applications in a production workload. Proc. ACM/IEEE Conf. on Supercomputing, p.388–397. http://dx.doi.org/10.1145/169627.169759

    Chapter  Google Scholar 

  • Plank, J.S., Beck, M., Kingsley, G., et al., 1995. Libckpt: transparent checkpointing under Unix. Proc. USENIX Technical Conf., p.18.

    Google Scholar 

  • Purakayastha, A., Ellis, C., Kotz, D., et al., 1995. Characterizing parallel file-access patterns on a large-scale multiprocessor. 9th Int. Parallel Processing Symp., p.165–172. http://dx.doi.org/10.1109/IPPS.1995.395928

    Chapter  Google Scholar 

  • Sisilli, J., 2015. Improved Solutions for I/O Provisioning and Application Acceleration. Available from http://www.flashmemorysummit.com/English/Collaterals/Proceedings/2015/20150811_FD11_Sisilli.pdf [Accessed on Nov. 18, 2015].

    Google Scholar 

  • Rudin, W., 1976. Principles of Mathematical Analysis. McGraw-Hill Publishing Co.

    MATH  Google Scholar 

  • Shalf, J., Dosanjh, S., Morrison, J., 2011. Exascale computing technology challenges. 9th Int. Conf. on High Performance Computing for Computational Science, p.1–25. http://dx.doi.org/10.1007/978-3-642-19328-6_1

    Google Scholar 

  • Strohmaier, E., Dongarra, J., Simon, H., et al., 2015. TOP500 Supercomputer Sites. Available from http://www.top500.org/ [Accessed on Dec. 30, 2015].

    Google Scholar 

  • Sun, X.H., Ni, L.M., 1993. Scalable problems and memorybounded speedup. J. Parall. Distr. Comput., 19(1):27–37. http://dx.doi.org/10.1006/jpdc.1993.1087

    Article  Google Scholar 

  • University of California, 2007. IOR HPC Benchmark. Available from http://sourceforge.net/projects/ior-sio/ [Accessed on Sept. 1, 2014].

    Google Scholar 

  • Wang, F., Xin, Q., Hong, B., et al., 2004. File system workload analysis for large scale scientific computing applications. Proc. 21st IEEE/12th NASA Goddard Conf. on Mass Storage Systems and Technologies, p.139–152.

    Google Scholar 

  • Wang, T., Oral, S., Wang, Y.D., et al., 2014. Burstmem: a high-performance burst buffer system for scientific applications. IEEE Int. Conf. on Big Data, p.71–79. http://dx.doi.org/10.1109/BigData.2014.7004215

    Google Scholar 

  • Wang, T., Oral, S., Pritchard, M., et al., 2015. Development of a burst buffer system for data-intensive applications. arXiv:1505.01765. Available from http://arxiv.org/abs/1505.01765

    Google Scholar 

  • Wang, Z.Y., 2009. Reliability speedup: an effective metric for parallel application with checkpointing. Int. Conf. on Parallel and Distributed Computing, Applications and Technologies, p.247–254. http://dx.doi.org/10.1109/PDCAT.2009.19

    Google Scholar 

  • Xie, B., Chase, J., Dillow, D., et al., 2012. Characterizing output bottlenecks in a supercomputer. Int. Conf. for High Performance Computing, Networking, Storage and Analysis, p.1–11. http://dx.doi.org/10.1109/SC.2012.28

    Google Scholar 

  • Yang, X.J., Du, J., Wang, Z.Y., 2011. An effective speedup metric for measuring productivity in large-scale parallel computer systems. J. Supercomput., 56(2):164–181. http://dx.doi.org/10.1007/s11227-009-0355-9

    Article  Google Scholar 

  • Yang, X.J., Wang, Z.Y., Xue, J.L., et al., 2012. The reliability wall for exascale supercomputing. IEEE Trans. Comput., 61(6):767–779. http://dx.doi.org/10.1109/TC.2011.106

    Article  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Wei Hu.

Additional information

Project supported by the National Natural Science Foundation of China (Nos. 61272141 and 61120106005) and the National High-Tech R&D Program (863) of China (No. 2012AA01A301)

ORCID: Wei HU, http://orcid.org/0000-0002-8839-7748

Rights and permissions

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hu, W., Liu, Gm., Li, Q. et al. Storage wall for exascale supercomputing. Frontiers Inf Technol Electronic Eng 17, 1154–1175 (2016). https://doi.org/10.1631/FITEE.1601336

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1631/FITEE.1601336

Key words

CLC number