High Performance Computing of Meshless Time Domain Method on Multi-GPU Cluster

Ikuno, Soichiro; Nakata, Satoshi; Hirokawa, Yuta; Itoh, Toshiya

doi:10.1088/1742-6596/574/1/012106

Cited by 1 publication

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Otros trabajos se centran en CPUs y GPUs que colaboran para resolver distintas partes del dominio de simulación de forma concurrente [7], y también se implementan entornos de cluster multi-GPU para conseguir mayor velocidad e investigar numéricamente el rendimiento [8], [9]. Se encuentran planteamientos MPI (message passing interface) -OpenMP [10]-GPU para calcular RCS con FDTD, que consiguen reducir el tiempo de cálculo de 3dias a 0.8h (RCS de un F-117) [11].…”

Section: Introductionunclassified

Nvidia CUDA parallel processing of large FDTD meshes in a desktop computer

Calatayud

Navarro-Modesto

Navarro

et al. 2020

Proceedings of the 10th Euro-American Conference on Telematics and Information Systems

View full text Add to dashboard Cite

The Finite Difference in Time Domain numerical (FDTD) method is a well know and mature technique in computational electrodynamics. Usually FDTD is used in the analysis of electromagnetic structures, and antennas. However still there is a high computational burden, which is a limitation for use in combination with optimization algorithms. The parallelization of FDTD to calculate in GPU is possible using Matlab and CUDA tools. For instance, the simulation of a planar array, with a three dimensional FDTD mesh 790x276x588, for 6200 time steps, takes one day -elapsed time-using the CPU of an Intel Core i3 at 2.4GHz in a personal computer, 8Gb RAM. This time is reduced 120 times when the calculation is parallelized and carried out in a Graphics Processing Unit (GPU) NVIDIA GeForce GTX 1080 Ti 11264 MB GDDR 5X. The elapsed time is reduced substantially, but also the simplicity of calculation and usefulness of a Matlab implementation. The elapsed time reduction is so substantial that the FDTD-Matlab-CUDA can be combined with optimization algorithms.

show abstract