Condensed forms for the symmetric eigenvalue problem on multi‐threaded architectures

Concurrency Computation Practice and Experience - Tập 23 Số 7 - Trang 694-707 - 2011
Paolo Bientinesi1, Francisco D. Igual2, Daniel Kreßner3, Matthias Petschow1, Enrique S. Quintana–Ort́ı2
1AICES, RWTH Aachen University, 52074 Aachen, Germany
2Depto. de Ingeniería y Ciencia de Computadores, Universidad Jaume I, 12071 Castellón, Spain#TAB#
3Seminar für Angewandte Mathematik ETH, Zurich, Switzerland

Tóm tắt

AbstractWe investigate the performance of the routines in LAPACK and the Successive Band Reduction (SBR) toolbox for the reduction of a dense matrix to tridiagonal form, a crucial preprocessing stage in the solution of the symmetric eigenvalue problem, on general‐purpose multi‐core processors. In response to the advances of hardware accelerators, we also modify the code in the SBR toolbox to accelerate the computation by off‐loading a significant part of the operations to a graphics processor (GPU). The performance results illustrate the parallelism and scalability of these algorithms on current high‐performance multi‐core and many‐core architectures. Copyright © 2010 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

Golub GH, 1996, Matrix Computations

Martin RM, 2008, Electronic Structure: Basic Theory and Practical Methods

10.1016/j.laa.2003.12.028

10.1137/030601107

Anderson E, 1992, LAPACK Users' Guide

10.1145/365723.365736

DongarraJ HammarlingSJ SorensenDC.Block reduction of matrices to condensed forms for eigenvalue computations. LAPACK Working Note 2 Technical Report MCS‐TM‐99 Argonne National Laboratory September 1987.

10.1016/S0167-8191(99)00021-6

Bientinesi P, 2009, Proceedings of the 8th International Conference on Parallel Processing and Applied Mathematics, PPAM 2009, 387

10.1137/0908009

10.1007/978-3-540-85451-7_79

10.1109/SC.2008.5214359

IgualFD Quintana‐OrtíG van de GeijnR.Level‐3 BLAS on a GPU: Picking the low hanging fruit. FLAME Working Note #37. DICC 2009‐04‐01 Universidad Jaume I. Depto ICC 2009.

Thông tin
Thông tin xuất bản

Concurrency Computation Practice and Experience

Tập 23 Số 7

694-707

Thông tin tác giả