Numerical reproducibility and parallel computations: Issues for interval algorithms

IEEE Transactions on Computers

Volumn 63, Issue 8, 2014, Pages 1915-1924

  Revol, Nathalie ^a   Theveny, Philippe ^b  

^a UNIVERSITÉ DE LYON   (France)

^b UNIV LYON   (France)

Author keywords

floating point arithmetic; Interval arithmetic; numerical reproducibility; parallel implementation; rounding mode

Indexed keywords

ALGORITHMS;

FLOATING-POINT ARITHMETIC; INTERVAL ARITHMETIC; PARALLEL IMPLEMENTATIONS; REPRODUCIBILITIES; ROUNDING MODES;

DIGITAL ARITHMETIC;

EID: 84904610736     PISSN: 00189340     EISSN: None     Source Type: Journal    
DOI: 10.1109/TC.2014.2322593     Document Type: Article

References (53)

1. D. H. Bailey, High-precision computation and reproducibility. presented at 2012 Workshop Reproducibility Comput. Exp. Math. [Online]. Available: http://www.davidhbailey.com/dhbtalks/dhb-icerm.pdf
2. K.-H. Brassel, "Advanced object-oriented technologies in modeling and simulation: The VSEit framework," in Proc. Eur. Simul. Multiconf., Prague, Czech Republic, 2001, pp. 154-160.
3. W.-F. Chiang, G. Gopalakrishnan, Z. Rakamaric, D. H. Ahn, and G. L. Lee, Determinism and reproducibility in large-scale HPC systems. presented at WODET 2013 - 4th Workshop Determinism Correctness Parallel Programm [Online]. Available: http://wodet.cs.washington.edu/wp-content/uploads/2013/03/wodet2013- final12.pdf
4. S. Foeml and J. F. Claerbout (editors), "Reproducible Research," Special issue of Comput. Sci. Eng., vol. 11, no. 1, 2009.
5. J. Demmel and H. D. Nguyen, Fast reproducible floating-point summation. presented at ARITH 2013 21st IEEE Symp. Comput. Arithmetic, pp. 163-172 [Online]. Available: http://www. eecs.berkeley.edu/~hdnguyen/public/papers/ ARITH21-Fast- Sum.pdf
6. J. Demmel and H. D. Nguyen, Numerical reproducibility and accuracy at exaScale. presented at ARITH 2013 21st IEEE Symp. Comput. Arithmetic [Online]. Available: http://www. arithsymposium.org/papers/paper-33.pdf
7. K. Diethelm, "The limits of reproducibility in numerical simulation," Comput. Sci. Eng., vol. 14, no. 1, pp. 64-71, 2012.
8. F. de Dinechin. (2013). De calculer juste a calculer au plus juste (in English except the title) School Accuracy Reproducibility Numer. Comput., Frejus, France [Online]. Available: http://calcul.math. cnrs.fr/IMG/pdf/2013- Dinechin-PRCN.pdf
9. J. Dongarra, Algorithmic and software challenges when moving towards exascale. presented at ISUM 2013: 4th Int. Supercomput. Conf. Mexico [Online]. Available: http://www.netlib.org/utk/people/JackDongarra/SLIDES/isum0213.pdf
10. D. Goldberg, "What every computer scientist should know about floating-point arithmetic," in ACM Comput. Surv., vol. 23, no. 1, pp. 5-48, 1991. (Pubitemid 21723680)
11. J. Gosling, "Java: An overview," Sun Microsystems Laboratories: The First Ten Years, 1991-2001 J. Treichel, M. Holzer, eds, Santa Clara, CA, USA:, Sun Microsystems, 1995.
12. J. Gosling, B. Joy, G. Steele, G. Bracha, and A. Buckley, The Java Language Specification, Java SE 7 Edition. Reading, MA, USA: Addison- Wesley, 2013.
13. W.D. Gropp, "Issues in accurate and reliable use of parallel computing in numerical programs," in Accuracy and Reliability in Scientific Computing, Bo Einarsson ed. Philadelphia, PA, USA: SIAM, 2005 pp. 253-263.
14. E. R. Hansen and G. W. Walster, Global Optimization Using Interval Analysis. 2nd ed. New York, NY, USA: Marcel Dekker, 2003.
15. Y. He and C. Q. Ding, "Using accurate arithmetics to improve numerical reproducibility and stability in parallel applications," J. Supercomput., vol. 18, pp. 259-277, 2001. (Pubitemid 32366268)
16. Y. Hida, X. S. Li, and D. H. Bailey. Quad double computation package, 2003-2012 [Online]. Available: http://crd-legacy.lbl. gov/~dhbailey/mpdist/
17. N. J. Higham, Accuracy and Stability of Numerical Algorithms. 2nd ed. Philadelphia, PA, USA: SIAM, 2002.
18. W. Hofschuster and W. Kramer. FI-LIB-A fast interval library (Version 1.2) in ANSI-C, 2005 [Online]. Available: http://www2. math.uni-wuppertal.de/ ~xsc/software/filib.html
19. IEEE Standard for Binary Floating-Point Arithmetic, ANSI/IEEE Standard, Std 754-1985, New York, NY, USA, 1985.
20. IEEE Standard for Floating-Point Arithmetic," IEEE Standard 754- 2008, New York, NY, USA, 2008.
21. W. Kahan and J. D. Darcy, How Java's floating-point hurts everyone everywhere presented at ACM 1998 Workshop Java High- Perform. Netw. Comput. [Online]. Available: http://www.cs.berkeley. edu/~wkahan/JAVAhurt.pdf
22. T. Kauranne, J. Oinonen, S. Saarinen, O. Serimaa, and J. Hietaniemi, "The operational HIRLAM 2 on parallel computers," in Proc. 6th Workshop Use Parallel Processors Meteorol., 1994, pp. 63-74.
23. B. Krammer, K. Bidmon, M. S. Muller, and M. M. Resch, "MARMOT: An MPI analysis and checking tool," Adv. Parallel Comput., vol. 13, pp. 493-500, 2004.
24. C. Lauter and V. Menissier-Morain, "There's no reliable computing without reliable access to rounding modes," in Proc. Symp. Sci. Comput., Comput. Arithmetics Verified Numerics, 2012, pp. 99-100.
25. M. Leeser, J. Ramachandran, T. Wahl, and D. Yablonski, OpenCL floating point software on heterogeneous architectures Portable or not? presented at 2012 Workshop Numerical Softw. Verification [Online]. Available: http://www.cs.rice.edu/~sc40/NSV/12/program.html
26. X. Li, J. Demmel, D. Bailey, G. Henry, Y. Hida, J. Iskandar, W. Kahan, A. Kapur, M. Martin, T. Tung, and D. J. Yoo, "Design, implementation and testing of extended and mixed precision BLAS," ACM Trans. Math. Softw., vol. 28, no. 2, pp. 152-205, 2002. (Pubitemid 135701674)
27. E. McIntosh, F. Schmidt, and F. de Dinechin, "Massive tracking on heterogeneous platforms," Proc. 9th Int. Comput. Accelerator Phys. Conf., 2006, Chamonix, France, pp. 13-16.
28. B. Marker, F. Van Zee, K. Goto, G. Quintana-Orti, and R. van de Geijn, "Toward scalable matrix multiply on multithreaded architectures," in Proc. 13th Int. Euro-Par Conf. Parallel Process., 2007, pp. 748-757.
29. J. K. Martinsen and H. Grahn, "A methodology for evaluating Java- Script execution behavior in interactive web applications," in Proc. 9th IEEE/ACS Int. Conf. Comput. Syst. Appl., 2011, pp. 241-248.
30. R. Moore, Interval Analysis. Englewood Cliffs, NJ, USA: Prentice Hall, 1966.
31. R. Moore, R. B. Kearfott, and M. J. Cloud, Introduction to Interval Analysis. Philadelphia, PA, USA: SIAM, 2009.
32. J.-M. Muller, N. Brisebarre, F. de Dinechin, C.-P. Jeannerod, V.Lefevre, G. Melquiond, N. Revol, and D. Stehle, S. Torres, Handbook of Floating-Point Arithmetic. New York, NY, USA: Birkhauser, 2010.
33. A. Neumaier, Interval Methods for Systems of Equations. Cambridge, U.K.: Cambridge Univ. Press, 1990.
34. H. D. Nguyen. (2011). Efficient algorithms for verified scientific computing: Numerical linear algebra using interval arithmetic, Ph.D. dissertation, ENS de Lyon [Online]. Available: http://tel.archives-ouvertes.fr/ tel-00680352/PDF/NGUYEN-Hong-Diep-2011-These.pdf
35. M. Philippsen, "Is Java ready for computational science?" in Proc. 2nd Eur. Parallel Distrib. Syst. Conf., Austria, 1998, pp. 299-304.
36. M. Philippsen, R. F. Boisvert, V. S. Getov, R. Pozo, J. Moreira, D. Gannon, and G. C. Fox, "JavaGrande - High performance computing with Java," in Proc. 5th Int. Workshop Appl. Parallel Comput., New Paradigms HPC Industry Academia, 2001, pp. 20-36. (Pubitemid 33239280)
37. D. Priest. (1997) Section D.11: Differences among IEEE 754 implementations, addendum to D. Goldberg: What every computer scientist should know about floating-point arithmetic [Online]. Available: https://web.archive. org/web/20010222141338/http://cch.loria.fr/documentation/IEEE754/ACM/addendum. html
38. N. Revol, Y. Denneulin, J.-F. Mehaut, and B. Planquelle, "Parallelization of continuous verified global optimization," in Proc. 19th IFIP TC7 Conf. Syst. Model. Optimization, Cambridge, U.K., 1999, http://perso.ens-lyon.fr/nathalie.revol/publis/RDMP99.pdf
39. N. Revol, K. Makino, and M. Berz, "Taylor models and floatingpoint arithmetic: Proof that arithmetic operations are validated in COSY," J. Logic Algebraic Programm., vol. 64, pp. 135-154, 2005. (Pubitemid 40556328)
40. N. Revol and P. Theveny, "Parallel implementation of interval matrix multiplication," Reliable Comput., vol. 19, no. 1, pp. 91-106, 2013.
41. N. Revol, "Tradeoffs between accuracy and efficiency for interval matrix multiplication," presented at Numerical Sofware: Design, Analysis and Verification, Santander, Spain, 2012..
42. R. W. Robey, J. M. Robey, and R. Aulwes, "In search of numerical consistency in parallel programming," Parallel Comput., vol. 37, pp. 217-229, 2011.
43. S. M. Rump, "INTLAB - INTerval LABoratory," in Developments in Reliable Computing, Tibor Csendes, ed., Norwell, MA, USA: Kluwer, 1999 pp. 77-104.
44. S. M. Rump, "Computer-assisted proofs and self-validating methods," in Accuracy and Reliability in Scientific Computing, Bo Einarsson, ed., Philadelphia, PA, USA: SIAM, 2005 pp. 195-240.
45. S. M. Rump, "Verification methods: Rigorous results using floating- point arithmetic," Acta Numerica, vol. 19, pp. 287-449, 2010.
46. S. M. Rump, "Fast interval matrix multiplication," Numerical Algorithms, vol. 61, no. 1, pp. 1-34, 2012.
47. V. Stodden, D. H. Bailey, J. Borwein, R. J. LeVeque, W. Rider, and W. Stein, Setting the default to reproducible - Reproducibility in computational and experimental mathematics. presented at 2012 Report on ICERM workshop Reproducibility Comput. Exp. Math., [Online]. Available: http://stodden.net/ icerm-report.pdf.
48. M. Taufer, O. Padron, P. Saponaro, and S. Patel, "Improving numerical reproducibility and stability in large-scale numerical simulations on GPUs," in Proc. IEEE Int. Symp. Parallel Distrib. Process., Atlanta, GA, USA, 2010, pp. 1-9.
49. G. K. Thiruvathukal, "Java at middle age: Enabling Java for computational science," Comput. Sci. Eng., vol. 2, no. 1, pp. 74-84, 2002.
50. F. Tisseur, "Newton's method in floating point arithmetic and iterative refinement of generalized eigenvalue problems," SIAM J. Matrix Anal. Appl., vol. 22, no. 4, pp. 1038-1057, 2001.
51. R. Todd. Introduction to conditional numerical reproducibility (CNR), 2012 [Online]. Available: http://software.intel.com/en-us/articles/introduction- to-the-conditional-numericalreproducibility- cnr
52. W. Tucker, Validated Numerics: A Short Introduction to Rigorous Computations. Princeton, NJ, USA: Princeton Univ. Press, 2011.
53. O. Villa, D. Chavarría Miranda, V. Gurumoorthi, A. Marquez, and S. Krishnamoorthy, Effects of floating-point non-associativity on numerical computations on massively multithreaded systems presented at 2009 Cray User Group Meeting [Online]. Available: https://cug.org/5-publications/proceedings- attendee-lists/CUG09CD/S09-Proceedings/pages/authors/01- 5Monday/4C-Villa/villa- paper.pdf