Accelerating scientific computations with mixed precision algorithms

Computer Physics Communications

Volumn 180, Issue 12, 2009, Pages 2526-2533

  Baboulin, Marc ^a   Buttari, Alfredo ^b   Dongarra, Jack ^c,d,e   Kurzak, Jakub ^c   Langou, Julie ^c   Langou, Julien ^f   Luszczek, Piotr ^g   Tomov, Stanimire ^c  

^a UNIVERSITY OF COIMBRA   (Portugal)

^b INRIA   (France)

^c University of Tennessee Knoxville   (United States)

^d OAK RIDGE NATIONAL LABORATORY   (United States)

^e UNIVERSITY OF MANCHESTER   (United Kingdom)

^f UNIVERSITY OF COLORADO   (United States)

^g MATHWORKS INC   (United States)

Author keywords

Iterative refinement; Mixed precision; Numerical linear algebra

Indexed keywords

BIT-OPERATIONS; CATALOGUE IDENTIFIERS; COEFFICIENT MATRIX; CONDITION NUMBERS; DISTRIBUTED PROGRAM; DOUBLE PRECISION; FLOATING-POINT ARITHMETIC; FORTRAN 77; GAUSSIAN ELIMINATION; GRAPHICAL PROCESSING UNITS; ILL-CONDITIONED; IRELAND; ITERATIVE PROCESS; ITERATIVE REFINEMENT; LINEAR ALGEBRA ALGORITHMS; LU FACTORIZATION; MIXED PRECISION; MODERN ARCHITECTURES; MODERN PROCESSORS; NUMERICAL ERRORS; NUMERICAL LINEAR ALGEBRA; NUMERICAL STABILITIES; OPERATING SYSTEMS; PERMUTATION MATRIX; PROGRAMMING LANGUAGE; ROUND-OFF ERRORS; RUNNING TIME; SCIENTIFIC COMPUTATION; SINGLE PRECISION; SOLUTION METHODS; SOLUTION OF LINEAR SYSTEMS; STI CELL; TEST DATA; TRIANGULAR MATRICES; UPPER TRIANGULAR MATRICES;

ALGEBRA; CELL MEMBRANES; COMPUTER ARCHITECTURE; COMPUTER OPERATING SYSTEMS; CONVERGENCE OF NUMERICAL METHODS; CORROSION PREVENTION; DIGITAL ARITHMETIC; ERROR CORRECTION; FACTORIZATION; FIELD PROGRAMMABLE GATE ARRAYS (FPGA); FORTRAN (PROGRAMMING LANGUAGE); LEARNING ALGORITHMS; LINEAR SYSTEMS; NUMBER THEORY; PROBLEM ORIENTED LANGUAGES; SEMICONDUCTING SILICON; SOLAR CONCENTRATORS; SPEECH ANALYSIS; TEST FACILITIES;

MATRIX ALGEBRA;

EID: 70350583245     PISSN: 00104655     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.cpc.2008.11.005     Document Type: Article

References (47)

1. Ypma T.J. SIAM Review 37 (1995) 531
2. Demmel J.W. Applied Numerical Linear Algebra (1997), SIAM
3. Wilkinson J.H. Rounding Errors in Algebraic Processes (1963), Prentice-Hall
4. Moler C.B. J. ACM 14 (1967) 316
5. Stewart G.W. Introduction to Matrix Computations (1973), Academic Press
6. Demmel J.W., et al. ACM Trans. Math. Software 32 (2006) 325
7. Oettli W., and Prager W. Numer. Math. 6 (1964) 405
8. Barrett R., et al. Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods (1994), Society for Industrial and Applied Mathematics, Philadelphia. http://www.netlib.org/templates/Templates.html Also available as postscript file at
9. Saad Y. Iterative Methods for Sparse Linear Systems (2003), Society for Industrial and Applied Mathematics, Philadelphia, PA, USA
10. Turner K., and Walker H.F. SIAM J. Sci. Stat. Comput. 13 (1992) 815
11. Axelsson O., and Vassilevski P.S. SIAM J. Matrix Anal. Appl. 12 (1991) 625
12. Golub G.H., and Ye Q. SIAM J. Scientific Comput. 21 (2000) 1305
13. Notay Y. SIAM J. Scientific Comput. 22 (2000) 1444
14. Y. Saad, A flexible inner-outer preconditioned GMRES algorithm, Technical Report 91-279, Department of Computer Science and Engineering, University of Minnesota, Minneapolis, Minnesota, 1991
15. Simoncini V., and Szyld D.B. SIAM J. Numer. Anal. 40 (2003) 2219
16. van den Eshof J., Sleijpen G.L.G., and van Gijzen M.B. J. Comput. Appl. Math. 177 (2005) 347
17. Vuik C. J. Comput. Appl. Math. 61 (1995) 189
18. van der Vorst H.A., and Vuik C. Numer. Linear Algebra Appl. 1 (1994) 369
19. Arioli M., Demmel J.W., and Duff I.S. SIAM J. Matrix Anal. Appl. 10 (1989) 165
20. Dongarra J.J., Croz J.D., Duff I.S., and Hammarling S. ACM Trans. Math. Software 16 (1990) 18
21. Dongarra J.J., Croz J.D., Duff I.S., and Hammarling S. ACM Trans. Math. Software 16 (1990) 1
22. Dongarra J.J., Croz J.D., Hammarling S., and Hanson R.J. ACM Trans. Math. Software 14 (1988) 18
23. Dongarra J.J., Croz J.D., Hammarling S., and Hanson R.J. ACM Trans. Math. Software 14 (1988) 1
24. Lawson C.L., Hanson R.J., Kincaid D., and Krogh F.T. ACM Trans. Math. Software 5 (1979) 308
25. Anderson E., et al. LAPACK Users' Guide. 3rd edition (1999), SIAM, Philadelphia
26. Buttari A., et al. Int. J. High Performance Comput. Appl. 21 (2007) 457
27. J. Langou, et al., Exploiting the performance of 32 bit floating point arithmetic in obtaining 64 bit accuracy, in: Proceedings of the 2006 ACM/IEEE Conference on Supercomputing, 2006
28. Kurzak J., Buttari A., and Dongarra J.J. IEEE Trans. Parallel Distrib. Systems 19 (2008) 1
29. Kurzak J., and Dongarra J.J. Concurrency Computat.: Pract. Exper. 19 (2007) 1371
30. Duff I.S., and Reid J.K. ACM Trans. Math. Software 9 (1983) 302
31. Ashcraft C., Grimes R., Lewis J., Peyton B.W., and Simon H. Intern. J. Supercomput. Appl. 1 (1987) 10
32. Buttari A., Dongarra J., Kurzak J., Luszczek P., and Tomov S. ACM Trans. Math. Software 34 (2008) 17 (Article 17, 22 pages)
33. Amestoy P.R., Duff I.S., and L'Excellent J.-Y. Comput. Methods Appl. Mech. Eng. 184 (2000) 501
34. Amestoy P.R., Duff I.S., L'Excellent J.-Y., and Koster J. SIAM J. Matrix Anal. Appl. 23 (2001) 15
35. Amestoy P.R., Guermouche A., L'Excellent J.-Y., and Pralet S. Parallel Comput. 32 (2006) 136
36. Davis T.A. ACM Trans. Math. Software 25 (1999) 1
37. Davis T.A. ACM Trans. Math. Software 30 (2004) 196
38. Davis T.A., and Duff I.S. SIAM J. Matrix Anal. Appl. 18 (1997) 140
39. Skeel R.D. Math. Comput. 35 (1980) 817
40. Higham N.J. Accuracy and Stability of Numerical Algorithms. 2nd edition (2002), SIAM
41. M. Arioli, I.S. Duff, Using FGMRES to obtain backward stability in mixed precision, in: Technical Report RAL-TR-2008-006, Rutherford Appleton Laboratory, 2008
42. Datta B.D. Numerical Linear Algebra and Applications (1995), Brooks Cole Publishing Company
43. K.O. Geddes, W.W. Zheng, Exploiting fast hardware floating point in high precision computation, in: Proceedings of the 2003 International Symposium on Symbolic and Algebraic Computation. Philadelphia, PA, USA, 2003, pp. 111-118
44. J.W. Demmel, Y. Hida, X.S. Li, E.J. Riedy, Extra-precise iterative refinement for overdetermined least squares problems, Technical Report EECS-2007-77, UC Berkeley, 2007, Also LAPACK Working Note 188
45. Björck A. Numerical Methods for Least Squares Problems (1996), SIAM
46. Dongarra J.J., Moler C.B., and Wilkinson J.H. SIAM J. Numer. Anal. 20 (1983) 23
47. Dongarra J.J. SIAM J. Scientific Statist. Comput. 4 (1983) 712