TRAP@NCI

Towards Ad-Hoc GPU Acceleration Of Parallel Eigensystem Computations

Garba, Michael T. and González-Vélez, Horacio (2011) Towards Ad-Hoc GPU Acceleration Of Parallel Eigensystem Computations. In: Proceedings 25th European Conference on Modelling and Simulation. ECMS, pp. 534-538. ISBN 9780956494429

[img]
Preview
PDF
Download (278kB) | Preview

Abstract

This paper explores the early implementation of high- performance routines for the solution of multiple large Hermitian eigenvector and eigenvalue systems on a Graphics Processing Unit (GPU). We report a perfor- mance increase of up to two orders of magnitude over the original EISPACK routines with a NVIDIA Tesla C2050 GPU, potentially allowing an order of magnitude in- crease in the complexity or resolution of a neutron scat- tering modeling application.

Item Type: Book Section
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Divisions: School of Computing > Staff Research and Publications
Related URLs:
Depositing User: CAOIMHE NI MHAICIN
Date Deposited: 29 Apr 2016 10:56
Last Modified: 29 Apr 2016 14:44
URI: http://trap.ncirl.ie/id/eprint/2184

Actions (login required)

View Item View Item