http://swrc.ontoware.org/ontology#Article
On Auto-tuned Pre/postprocessing for the Singular Value Decomposition of Dense Square Matrices
en
自動チューニング
NS Solutions Corporation
Graduate School of Informatics, Kyoto University
Graduate School of System Informatics, Kobe University
Graduate School of Humanities and Sciences, Nara Women’s University
Graduate School of Informatics, Kyoto University
Graduate School of Informatics, Kyoto University
Hiroki Toyokawa
Kinji Kimura
Yusaku Yamamoto
Masami Takata
Akira Ajisaka
Yoshimasa Nakamura
An auto-tuning technique is devised for fast pre/postprocessing for the singular value decomposition of dense square matrices with the Dongarra or the Bischof-Murata algorithms. The computation speed of these two algorithms varies depending on a parameter and specification of computers. By dividing these algorithms into several parts and by modeling each of them, we can estimate their computation times accurately. This enables us to choose an optimal parameter and the faster algorithm prior to execution. Consequently the pre/postprocessing is done faster and the singular value decomposition is applied faster to dense square matrices. Numerical experiments show the effectiveness of the proposed auto-tuning function. The I-SVD library, which incorporates this auto-tuning function, has been published.
An auto-tuning technique is devised for fast pre/postprocessing for the singular value decomposition of dense square matrices with the Dongarra or the Bischof-Murata algorithms. The computation speed of these two algorithms varies depending on a parameter and specification of computers. By dividing these algorithms into several parts and by modeling each of them, we can estimate their computation times accurately. This enables us to choose an optimal parameter and the faster algorithm prior to execution. Consequently the pre/postprocessing is done faster and the singular value decomposition is applied faster to dense square matrices. Numerical experiments show the effectiveness of the proposed auto-tuning function. The I-SVD library, which incorporates this auto-tuning function, has been published.
AA11833852
情報処理学会論文誌コンピューティングシステム（ACS）
4
3
9-21
2011-05-12
1882-7829