Abstract
The methods are shown for overcoming rounding errors on different kinds of parallel computers. The typical problems are demonstrated and situation are investigated which lead to large increase of the errors mentioned. A family of packages are described for organizing the super-precision computations with dynamically changed length of operands for both SIMD and MIMD models of computations.
Preview
Unable to display preview. Download preview PDF.
References
Aguliar M.F., Duprat J. 'Towards a high precision massively parallel computer' Proc. of PARLE'91, Parallel Architectures and Languages Europe, Athens, Greece, Springer-Verlag, Berlin, P. 73–84, 1994. (Lect. Not. in Comp. Sci., 817).
Batcher, K.E. Bit STARAN series E, Proc. of Int. Conf. Parallel Processing IEEE, New York, pp. 140–143, 1977.
Bleher J.H., Rump S.M., Kulisch U., et al. 'FORTRAN-SC: a Study of a FORTRAN Extension for Engineering/ Scientific Computation with Access to ACRITH', Computing, Springer, Vol. 39, P. 93–110, 1987.
Buell D., Ward R. 'A multiprecise integer arithmetic package' The Journal of Supercomputing, Vol. 3, N 2, P. 89–107, 1989.
Davidenkof A. ‘High accuracy arithmetic on transputers’ Computer Arithmetic, Scientific Computation and Mathematical Modelling. Proc. of SCAN'90. Albena, 1990. IMACS Annals on Computing and Applied Mathematic, Vol. 12, (1992), J.C. Baltzer AG, Basel, P. 45–61, 1991.
De Pinto P., Lauria Fr. E., Sette M. `Implementing arbitrary precision arithmetic in a general purpose neural networks' Parallel Computing: Problems, Methods and Applications, Elsevier Science Publ. B.V., P. 477–485, 1992.
Matte R., Kulisch U., Neaga M., Ratz D., Ullrich Ch. Pascal-XSC. Language Reference with Examples Springer-Verlag 1991. 5Vol.C-36, N 9, P. 1088–1096, 1987.
Michelucci D. ‘Arithmetic Issues in Geometric Computations’ Proc. Second Real Numbers and Computer Conference pages 43–69, CIRM, Marseille, France, April 1996.
D. Ratz, D. ‘The effects of the arithmetic of vector computers on basic numerical methods’ Contributions to Computer Arithmetic and Self-Validating numerical methods. IMACS Annals on Computing and Applied Mathematics, J.C.Baltzer AG, Basel, P. 499–514, 1990.
Vazhenin, A. ‘Programming system for high-accuracy computations for associative array processors’ Proc. of the Int. Conf. CONPAR 90-VAPP IV, Vol. of Special Technical Contributions, Zürich, 1990, pp. 69–72.
Vazhenin A.P., Mirenkov N.N. 'SCORE: Scientific Computers Overcoming Rounding Errors' PARALLEL COMPUTING: Trends and Applications, North-Holland, P. 387–394, 1994.
Vazhenin A. 'Efficient high-accuracy computations in massively parallel systems' Proc. “Workshop on Parallel Scientific Computing PARA94-L”, Lyngby, Denmark, Springer-Verlag, Berlin, 1994, P. 505–519. (Lect. Not. in Comp. Sci., 879).
Vazhenin A., Morozov V. Parallel iterative solution of linear equations with dynamically changed length of operands, Proc. of the Int. Conf. Parallel Computing Technilogies (PaCT-95), 1995, Springer-Verlag, Lect. Notes in Comp. Sci., Vol.964, pp. 294–303.
Vazhenin A., Morozov V. Portable Multiprecision Arithmetic Package Based on Message Passing Interface, Proc. of the International Conference “Parallel Algorithms/Architecture Sysnthesis” (pAs'97), 1997, IEEE Computer Society Press, Los Alamitos, California, pp. 324–331.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Vazhenin, A., Morozov, V. (1997). SPARTH: A family of systems for parallel high-accuracy computations. In: Malyshkin, V. (eds) Parallel Computing Technologies. PaCT 1997. Lecture Notes in Computer Science, vol 1277. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63371-5_53
Download citation
DOI: https://doi.org/10.1007/3-540-63371-5_53
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63371-6
Online ISBN: 978-3-540-69525-7
eBook Packages: Springer Book Archive