Acoustic scattering solver based on single level FMM for multi-GPU systems

López-Portugués, Miguel; López-Fernández, Jesús A.; MenéNdez-Canal, Jonatan; Rodríguez-Campa, Alberto; Ranilla, José

doi:10.1016/j.jpdc.2011.07.013

Cited by 9 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the FMM is implemented in a code based on the sequential use of the Central Processing Unit (CPU) for a system with N unknowns, the computing times is considerably reduced, with times O N 1.5 , or even O N log 2 N with a multilevel approach (see López-Portugués et al, 2012 ).…”

Section: Introductionmentioning

confidence: 99%

An Iterative Parallel Solver in GPU Applied to Frequency Domain Linear Water Wave Problems by the Boundary Element Method

Molina

Martínez-Castro

Ortiz

2018

Front. Built Environ.

View full text Add to dashboard Cite

In this paper a parallel iterative solver based on the Generalized Minimum Residual Method (GMRES) with complex-valued coefficients is explored, with applications to the Boundary Element Method (BEM). The solver is designed to be executed in a GPU (Graphic Processing Unit) device, exploiting its massively parallel capabilities. The BEM is a competitive method in terms of reduction in the number of degrees of freedom. Nonetheless, the BEM shows disadvantages when the dimension of the system grows, due to the particular structure of the system matrix. With difference to other acceleration techniques, the main objective of the proposed solver is the direct acceleration of existing standard BEM codes, by transfering to the GPU the solver task. The CUDA programming language is used, exploiting the particular architecture of the GPU device for complex-valued systems. To explore the performances of the solver, two linear water wave problems have been tested: the frequency-dependent added mass and damping matrices of a 3D floating body, and the Helmholtz equation in a 2D domain. A NVidia GeForce GTX 1080 graphic card has been used. The parallelized GMRES solver shows reductions in computing times when compared with its CPU implementation.

show abstract

Section: Introductionmentioning

confidence: 99%

An Iterative Parallel Solver in GPU Applied to Frequency Domain Linear Water Wave Problems by the Boundary Element Method

Molina

Martínez-Castro

Ortiz

2018

Front. Built Environ.

View full text Add to dashboard Cite

show abstract

“…Taking into account that the FMM has been efficiently implemented in Graphics Processing Units (GPUs) for efficient analysis of electromagnetic scattering problems [13] and acoustics [14], it is proposed to follow a similar methodology for the inverse operator, the IFMM, so that the computationally intensive parts of the IFMM code can be evaluated to the same accuracy at higher speeds by using a GPU. Parallelization using GPUs is desired for the IFMM algorithm given that it can be used as an image reconstruction algorithm in a mm-wave concealed threat detection system.…”

Section: Introductionmentioning

confidence: 99%

A Gpu Implementation of the Inverse Fast Multipole Method for Multi-Bistatic Imaging Applications

Tirado¹,

Ghazi²,

Álvarez-López³

et al. 2017

PIER M

View full text Add to dashboard Cite

Abstract-This paper describes a parallel implementation of the Inverse Fast Multipole Method (IFMM) for multi-bistatic imaging configurations. NVIDIAs Compute Unified Device Architecture (CUDA) is used to parallelize and accelerate the imaging algorithm in a Graphics Processing Unit (GPU). The algorithm is validated with synthetic data generated by the Modified Equivalent Current Approximation (MECA) method and experimental data collected by a Frequency-Modulated Continuous Wave (FMCW) radar system operating in the 70-77 GHz frequency band. The presented results show that the IFMM implementation using the CUDA platform is effective at significantly reducing the algorithm computational time, providing a 300X speedup when compared to the single core OpenMP version of the algorithm.

show abstract

“…The first contribution, by López-Portugués et al [1], presents a parallel solver of a high frequency single level Fast Multipole Method (FMM) for the Helmholtz equation applied to acoustic scattering. Their application exploits both the CPUs and the GPUs of a workstation, resulting in a heterogeneous solution that achieves runtimes comparable to those of a multicore cluster while reducing the power consumption by more than an order of magnitude.…”

mentioning

confidence: 99%