Abstract
Due to increasing complexity of supercomputers, hard and soft errors are causing more and more problems in high-performance scientific and engineering computation. In order to improve reliability (increase the mean time to failure) of computing systems, a lot of efforts have been devoted to developing techniques to forecast, prevent, and recover from errors at different levels, including architecture, application, and algorithm. In this paper, we focus on algorithmic error resilient iterative solvers and introduce a redundant subspace correction method. Using a general framework of redundant subspace corrections, we construct iterative methods, which have the following properties: (1) maintain convergence when error occurs assuming it is detectable; (2) introduce low computational overhead when no error occurs; (3) require only small amount of point-to-point communication compared to traditional methods and maintain good load balance; (4) improve the mean time to failure. Preliminary numerical experiments demonstrate the efficiency and effectiveness of the new subspace correction method. For simplicity, the main ideas of the proposed framework were demonstrated using the Schwarz methods without a coarse space, which do not scale well in practice.
Similar content being viewed by others
Notes
The y-axis is processing units and the x-axis is time. The solid bars stand for computational work and springs stand for inter-process communication.
References
Abts, D., Thompson, J., Schwoerer, G.: Architectural Support for Mitigating Dram Soft Errors in Large-Scale Supercomputers. Tech. rep. (2006)
Bjorstad, P.E., Skogen, M.: Domain decomposition algorithms of schwarz type, designed for massively parallel computers. In: 5th International Symposium on Domain Decomposition Methods for Partial Differential Equations. SIAM, Philadelphia, pp. 362–375 (1992)
Boley, D.L., Brent, R.P., Golub, G.H., Luk, F.T.: Algorithmic fault tolerance using the lanczos method. SIAM J. Matrix Anal. Appl. 13(1), 312–332 (1992)
Bramble, J.H., Pasciak, J.E., Xu, J.: Parallel multilevel preconditioners. Math. Comput. 55(191), 1–22 (1990)
Bronevetsky, G., de Supinski, B.R.: Soft error vulnerability of iterative linear algebra methods. In: Proceedings of the 22nd Annual International Conference on Supercomputing, pp. 155–164 (2008)
Chen, Z., Dongarra, J.: Algorithm-based fault tolerance for fail-stop failures. IEEE Trans. Parallel Distrib. Syst. 19(12), 1628–1641 (2008)
Deng, Y.: Applied Parallel Computing. World Scientific, Singapore (2013)
Dongarra, J., Beckman, P., Moore, T., Aerts, P., Aloisio, G., Andre, J.C., Barkai, D., Berthou, J.Y., Boku, T., Braunschweig, B., Cappello, F., Chapman, B.: Choudhary, a., Dosanjh, S., Dunning, T., Fiore, S., Geist, a., Gropp, B., Harrison, R., Hereld, M., Heroux, M., Hoisie, a., Hotta, K., Ishikawa, Y., Johnson, F., Kale, S., Kenway, R., Keyes, D., Kramer, B., Labarta, J., Lichnewsky, a., Lippert, T., Lucas, B., Maccabe, B., Matsuoka, S., Messina, P., Michielse, P., Mohr, B., Mueller, M.S., Nagel, W.E., Nakashima, H., Papka, M.E., Reed, D., Sato, M., Seidel, E., Shalf, J., Skinner, D., Snir, M., Sterling, T., Stevens, R., Streitz, F., Sugar, B., Sumimoto, S., Tang, W., Taylor, J., Thakur, R., Trefethen, a., Valero, M., van der Steen, a., Vetter, J., Williams, P., Wisniewski, R., Yelick, K.: The international exascale software project roadmap. Int. J. High Perform. Comput. Appl. 25(1), 3–60 (2011). doi:10.1177/1094342010391989
Dryja, M., Widlund, O.: Some domain decomposition algorithms for elliptic problems. In: Hayes, L., Kincaid, D. (eds.) Iterative Methods for Large Linear Systems, pp. 273–291. Academic Press, San Diego (1989)
Dryja, M., Widlund, O.B.: Additive schwarz methods for elliptic finite element problems in three dimensions. In: Fifth Conference on Domain Decomposition Methods for Partial Differential Equations, Philadelphia, PA (1992)
Du, P., Luszczek, P., Dongarra, J.: High performance dense linear system solver with resilience to multiple soft errors. In: International Conference on Cluster Computing, pp. 272–280 (2011)
Du, P., Luszczek, P., Dongarra, J.: High performance dense linear system solver with resilience to multiple soft errors. Procedia Comput. Sci. 9, 216–225 (2012). doi:10.1016/j.procs.2012.04.023
Gropp, W.D.: Parallel computing and domain decomposition. In: Fifth Conference on Domain Decomposition Methods for Partial Differential Equations, pp. 349–361 (1992)
Hackbusch, W.: Elliptic Differential Equations: Theory and Numerical Treatment, Computational Mathematics Series. Springer, Berlin (1992)
Hackbusch, W.: Iterative Solution of Large Sparse Systems of Equations, Applied Mathematical Sciences, vol. 95. Springer, New York (1994)
Hoemmen, M., Heroux, M.A.: Fault-tolerant iterative methods via selective reliability. In: Proceedings of the 2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) (2011)
Huang, K.h., Abraham, J.A.: Algorithm-based fault tolerance for matrix operations. IEEE Trans. Comput. c(6), 518–528 (1984)
Karypis, G., Kumar, V.: A fast and high quality multilevel scheme for partitioning irregular graphs. SIAM J. Sci. Comput. 20(1), 359–392 (1998)
Keyes, D.E.: Exaflop/s: the why and the how. Comptes Rendus Mécanique 339(2–3), 70–77 (2011). doi:10.1016/j.crme.2010.11.002
Kikuchi, N.: Finite Element Methods in Mechanics. Cambridge University Press, Cambridge (1986)
Langou, J., Chen, Z., Bosilca, G., Dongarra, J.: Recovery patterns for iterative methods in a parallel unstable environment. SIAM J. Sci. Comput. 30(1), 102–116 (2007)
Laprie, J.: Dependable computing: Concepts, limits, challenges. In: The 25th IEEE International Symposium on Fault-Tolerant Computing, pp. 42–54 (1995)
Liu, Y., Nassar, R., Leangsuksun, C.B., Naksinehaboon, N., Paun, M., Scott, S.L.: An optimal checkpoint/restart model for a large scale high performance computing system. In: 2008 IEEE International Symposium on Parallel and Distributed Processing, pp. 1–9 (2008). doi:10.1109/IPDPS.2008.4536279
Luk, F., Park, H.: An analysis of algorithm-based fault tolerance techniques. In: 30th Annual Technical Symposium on International Society for Optics and Photonics, pp. 172–184 (1986)
Malkowski, K., Raghavan, P., Kandemir, M.: Analyzing the soft error resilience of linear solvers on multicore multiprocessors. In: 2010 IEEE International Symposium on Parallel & Distributed Processing, pp. 1–12 (2010). doi:10.1109/IPDPS.2010.5470411
Michalak, S., Harris, K., Hengartner, N., Takala, B., Wender, S.: Predicting the number of fatal soft errors in Los Alamos national laboratory’s ASC Q supercomputer. IEEE Trans. Device Mater. Reliab. 5(3), 329–335 (2005). doi:10.1109/TDMR.2005.855685
Miskov-Zivanov, N., Marculescu, D.: Soft error rate analysis for sequential circuits. In: Proceedings of the Conference on Design, Automation and Test in Europe, pp. 1436–1441 (2007)
Monk, P.: Finite Element Methods for Maxwell’s Equations. Numerical Mathematics and Scientific Computation. Clarendon Press, Oxford (2003)
Mukherjee, S., Emer, J., Reinhardt, S.K.: The soft error problem: An architectural perspective. In: Proc. 11th Int’l Symp. on High-Performance Computer Architecture (HPCA) (2005)
PHG (Parallel Hierarchical Grid). http://lsec.cc.ac.cn/phg/
Plank, J.S., Li, K., Puening, M.A.: Diskless checkpointing. IEEE Trans. Parallel Distrib. Syst. 9(10), 972–986 (1998)
Reddi, V.: Hardware and software co-design for robust and resilient execution. In: 2012 International Conference on Collaboration Technologies and Systems, p. 380 (2012)
Roy-Chowdhury, A., Banerjee, P.: A fault-tolerant parallel algorithm for iterative solution of the laplace equation. In: International Conference on Parallel Processing, vol. 3, pp. 133–140 (1993)
Saad, Y.: Iterative Methods for Sparse Linear Systems. SIAM (1996). doi:10.1109/MCSE.1996.1231631
Shantharam, M., Srinivasmurthy, S., Raghavan, P.: Characterizing the impact of soft errors on iterative methods in scientific computing. In: Proceedings of the International Conference on Supercomputing, pp. 152–161 (2011)
Shantharam, M., Srinivasmurthy, S., Raghavan, P.: Fault tolerant preconditioned conjugate gradient for sparse linear system solution. In: Proceedings of the 26th ACM International Conference on Supercomputing, pp. 69–78. ACM, New York (2012)
Smith, B.F.: A parallel implementation of an iterative substructuring algorithm for problems in three dimensions. SIAM J. Sci. Comput. 14(2), 406–423 (1993)
Stoyanov, M.K., Webster, C.G.: Numerical Analysis of Fixed Point Algorithms in the Presence of Hardware Faults. Tech. rep., Oak Ridge National Laboratory (ORNL) (2013)
Toselli, A., Widlund, O.B.: Domain Decomposition Methods: Algorithms and Theory, Springer Series in Computational Mathematics, vol. 34. Springer, Berlin (2005)
Treaster, M.: A Survey of Fault-tolerance and Fault-recovery techniques in Parallel Systems. Tech. rep., ACM Computing Research Repository (2005)
Xu, J.: Iterative methods by space decomposition and subspace correction. SIAM Rev. 34, 581–613 (1992)
Xu, J., Zikatanov, L.: The method of alternating projections and the method of subspace corrections in Hilbert space. J. Am. Math. Soc. 15(3), 573–597 (2002). doi:10.1090/S0894-0347-02-00398-3
Zhang, W.: Computing cache vulnerability to transient errors and its implication. In: 20th IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems, pp. 427–435 (2005)
Acknowledgements
Cui and Zhang are partially supported by National Key Research and Development Program 2016YFB0201304, by China NSF Grants 91430215 and 91530323, and by National Center for Mathematics and Interdisciplinary Sciences of Chinese Academy of Sciences (NCMIS). Xu is partially supported by NSF DMS-0915153 and DOE DE-SC0006903.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Gabriel Wittum.
Rights and permissions
About this article
Cite this article
Cui, T., Xu, J. & Zhang, CS. An error-resilient redundant subspace correction method. Comput. Visual Sci. 18, 65–77 (2017). https://doi.org/10.1007/s00791-016-0270-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00791-016-0270-6