Abstract
Three parallel algorithms for the eigensolution of real symmetric matrices of order n on a SIMD-type parallel computer with an associative memory are considered. The algorithms realize various parallel orderings of the Jacobi orthogonalization procedure. A detailed description of the parallel computational process is given which allows the power of the machine to be exploited. The arithmetic parallel complexity achieved for the complete solution is 0(n), the number of parallel data transfers is 0(n log n). The results of algorithm simulations are included.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
Brent, R.P., Luk, F.T., A systolic architecture for the singular value decomposition. Tech. Rep. TR-CS-92-09, Aust.Nat. Univ. (1982).
Modi, J.J., Pryce, J.D., Efficient implementation of Jacobi's diagonalization method on the DAP. Numer. Math. 46, 3 (1985), 443–454.
Richter, K., Parallel computer system SIMD. In: Artif. Intelligence and Inf.-Control Syst. of Robots, I. Plander, Ed., North-Holland, Amsterdam, 1984.
Rutishauser, H., The Jacobi method for real symmetric matrices. In: Handbook for Automatic Computation 2, J.H. Wilkinson and C. Reinsch, Eds., Springer-Verlag, Berlin, 1971.
Sameh, A.H., On Jacobi and Jacobi-like algorithms for a parallel computer. Math. Comput. 25 (1971), 579–590.
Vajteršic, M., Some linear algebra algorithms proposed for a parallel associative computer. In: Proc. Algorithms 85, JSMF Bratisava, 1985.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1986 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Vajteršic, M. (1986). Fast parallel algorithms for eigenvalue and singular value computations. In: Händler, W., Haupt, D., Jeltsch, R., Juling, W., Lange, O. (eds) CONPAR 86. CONPAR 1986. Lecture Notes in Computer Science, vol 237. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-16811-7_178
Download citation
DOI: https://doi.org/10.1007/3-540-16811-7_178
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-16811-9
Online ISBN: 978-3-540-44856-3
eBook Packages: Springer Book Archive