×

SVD computations on the connection machine CM-5/CM-5E: Implementation and accuracy. (English) Zbl 0891.65037

A technique for obtaining the singular value decomposition (SVD) of a real dense matrix is to reduce the dense matrix to bidiagonal form and then compute the SVD of the resulting matrix. This approach is efficiently implemented on the Connection Machine CM-5/CM-5E. Timing results are given, and numerical experiments (with bidiagonal matrices) are presented.

MSC:

65F20 Numerical solutions to overdetermined systems, pseudoinverses
65F15 Numerical computation of eigenvalues and eigenvectors of matrices
65Y05 Parallel numerical computation
PDFBibTeX XMLCite
Full Text: DOI