Iterative solution of linear systems in electromagnetics (and not only): experiences with CUDA

Authors:
Danilo De Donno;Alessandra Esposito;Giuseppina Monti;Luciano Tarricone
Affiliations:
Department of Innovation Engineering, University of Salento, Lecce, Italy;Department of Innovation Engineering, University of Salento, Lecce, Italy;Department of Innovation Engineering, University of Salento, Lecce, Italy;Department of Innovation Engineering, University of Salento, Lecce, Italy
Venue:
Euro-Par 2010 Proceedings of the 2010 conference on Parallel processing
Year:
2010

Citing 2
Cited 1

Field Computation by Moment Methods

Field Computation by Moment Methods
Concurrent number cruncher: a GPU implementation of a general sparse linear solver

International Journal of Parallel, Emergent and Distributed Systems

The BiConjugate gradient method on GPUs

The Journal of Supercomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose the use of graphics processing units as a low-cost and efficient solution of electromagnetic (and other) numerical problems. Based on the software platform CUDA (Compute Unified Device Architecture), a solver for unstructured sparse matrices with double precision complex data has been implemented and tested for several practical cases. Benchmark results confirm the validity of the proposed software in terms of speed-up, speed and GPU execution time.