標題: A Note on the GPU Acceleration of Eigenvalue Computations
作者: Rupp, K.
Tillet, Ph
Smith, B. F.
Grasser, T.
Juengel, A.
電機工程學系
Department of Electrical and Computer Engineering
關鍵字: Linear Algebra;Eigenvalues;Krylov methods;GPU;CUDA;OpenCL;ViennaCL;PETSc;SLEPc
公開日期: 2013
摘要: Eigenvalue computations for large sparse matrices such as the Lanczos method are commonly based on Krylov subspace techniques. One of the dominant operations in such algorithms are iterated computations of inner products with the same vector in order to preserve orthogonality of the Krylov basis. These operations can be accelerated by existing BLAS functionality using GPUs. However, this is not fully efficient due to unnecessary memory transfers. We present improved implementations in CUDA and OpenCL, which are now available in ViennaCL, PETSc and SLEPc, and demonstrate an up to two-fold performance gain over existing GPU vendor libraries.
URI: http://dx.doi.org/10.1063/1.4825816
http://hdl.handle.net/11536/135393
ISBN: 978-0-7354-1185-2
ISSN: 0094-243X
DOI: 10.1063/1.4825816
期刊: 11TH INTERNATIONAL CONFERENCE OF NUMERICAL ANALYSIS AND APPLIED MATHEMATICS 2013, PTS 1 AND 2 (ICNAAM 2013)
Volume: 1558
起始頁: 1536
結束頁: 1539
顯示於類別:會議論文