Matrix inversion on CPU-GPU platforms with applications in control theory / Peter Benner, Pablo Ezzatti, Enrique S. Quintana-Ortí, Alfredo Remón

Anzeigen / Download258.79 KB

Discovery

870315951

URN

urn:nbn:de:gbv:3:2-63936

DOI

ISBN

ISSN

Beiträger

Erschienen

Magdeburg : Max Planck Institute for Dynamics of Complex Technical Systems, February 1, 2012

Umfang

1 Online-Ressource (18 Seiten = 0,25 MB) : Diagramme

Ausgabevermerk

Sprache

eng

Anmerkungen

Inhaltliche Zusammenfassung

Abstract: In this paper we tackle the inversion of large-scale dense matrices via conventional matrix factorizations (LU, Cholesky, LDLT ) and the Gauss-Jordan method on hybrid platforms consisting of a multi-core CPU and a many-core graphics processor (GPU). Specifically, we introduce the different matrix inversion algorithms using a unified framework based on the notation from the FLAME project; we develop hybrid implementations for those matrix operations underlying the algorithms, alternative to those in existing libraries for single-GPU systems; and we perform an extensive experimental study on a platform equipped with state-of-the-art general-purpose architectures from Intel and a “Fermi” GPU from NVIDIA that exposes the efficiency of the different inversion approaches. Our study and experimental results show the simplicity and performance advantage of the GJE-based inversion methods, and the difficulties associated with the symmetric indefinite case.

Schriftenreihe

Max Planck Institute Magdeburg Preprints ; 12-02 ppn:870173030

Gesamttitel

Band

Zeitschriftentitel

Bandtitel

Beschreibung

Schlagwörter

Zitierform

enthaltene Monographien

enthalten in mehrteiligem Werk

Vorgänger dieser Zeitschrift

Nachfolger dieser Zeitschrift