Development of a CUDA Implementation of the 3D FDTD Method

Livesey, Matthew; Stack, James F.; Costen, Fumie; Nanri, Takeshi; Nakashima, Naotoshi; Fujino, Seiji

doi:10.1109/map.2012.6348145

Cited by 42 publications

(19 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Successful hardware accelerations for large-scale FDTD simulation using GPUs have been reported in the literature (see, for example, [2][3][4][5][6][7][8][9][10][11][12][13]). Compared with CPU computing, GPU computing offers a high-performance leap at a very reasonable cost.…”

Section: Introductionmentioning

confidence: 99%

An FDTD-Based Parallel Virtual Tool for RCS Calculations of Complex Targets

Çakır

Sevgi

2014

IEEE Antennas Propag. Mag.

View full text Add to dashboard Cite

A novel, accelerated, and parallelized Finite-Difference Time-Domain (FDTD) based radar cross section (RCS) prediction tool, MGL-FastRCS, is introduced. MGL-FastRCS has been parallelized with OpenMP and Compute Unifi ed Device Architecture (CUDA) for high-performance computing with multiple cores and graphics processing units (GPUs). Characteristic examples using realistic complex air and surface targets are presented. The virtual tool automatically constructs a discrete wire-grid model of the structure under investigation. It is therefore also possible to compare the FDTD results against Method-of-Moments (MoM) data.

show abstract

Section: Introductionmentioning

confidence: 99%

An FDTD-Based Parallel Virtual Tool for RCS Calculations of Complex Targets

Çakır

Sevgi

2014

IEEE Antennas Propag. Mag.

View full text Add to dashboard Cite

show abstract

“…There are two major approaches to improve the computational efficiency of the FDTD method, developing either hardware or algorithmic methodology. The hardware approach involves the parallelization of the computation by 1) using multiple cores in shared and/or distributed memory architectures (or graphics processing units [2]) and 2) taking advantage of modern processors' features such as the register level parallelization, i.e., the streaming SIMD extensions (SSEs) [3] or the advanced vector extensions (AVXs).…”

Section: Efficient Parallel Lod-fdtd Methods Formentioning

confidence: 99%

Efficient Parallel LOD-FDTD Method for Debye-Dispersive Media

Hemmi

Costen

Garcia

et al. 2014

IEEE Trans. Antennas Propagat.

View full text Add to dashboard Cite

The locally one-dimensional finite-difference time-domain (LOD-FDTD) method is a promising implicit technique for solving Maxwell's equations in numerical electromagnetics. This paper describes an efficient message passing interface (MPI)-parallel implementation of the LOD-FDTD method for Debye-dispersive media. Its computational efficiency is demonstrated to be superior to that of the parallel ADI-FDTD method. We demonstrate the effectiveness of the proposed parallel algorithm in the simulation of a bio-electromagnetic problem: the deep brain stimulation (DBS) in the human body.Index Terms-Distributed memory systems, electromagnetic fields, electromagnetic propagation in dispersive media, finite-difference methods, numerical analysis, parallel programming, time-domain analysis.

show abstract

“…The main drawback of the numerical methods in general and particularly of the methods mentioned above is that computational costs and especially solution time increase substantially when a structure under analysis becomes more complicated. Therefore, it is important to develop new approaches to speed up such calculations, both via hardware [12] and algorithms.…”

Section: Approaches To Solving the Sequence Of Linear Systemsmentioning

confidence: 99%

Choosing order of operations to accelerate strip structure analysis in parameter range

Kuksenko

Akhunov

Gazizov

2018

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Abstract. The paper considers the issue of using iteration methods in solving the sequence of linear algebraic systems obtained in quasistatic analysis of strip structures with the method of moments. Using the analysis of 4 strip structures, the authors have proved that additional acceleration (up to 2.21 times) of the iterative process can be obtained during the process of solving linear systems repeatedly by means of choosing a proper order of operations and a preconditioner. The obtained results can be used to accelerate the process of computer-aided design of various strip structures. The choice of the order of operations to accelerate the process is quite simple, universal and could be used not only for strip structure analysis but also for a wide range of computational problems. IntroductionDistributed circuits based on various strip structures are widely used in radioelectronic equipment, both as transmission lines that maintain proper characteristics for desired signals for a long time and as a basis for new protective devices. A strip structure consists of signal and ground conductors and a dielectric substrate. The separation between conductors, their thickness, other geometrical parameters and dielectric permittivity of a substrate can be repeatedly changed during simulation and optimization of elements and devices. These processes significantly increase computational costs. Generally, hardware accelerators (multicore workstations, clusters, graphical processing units) are used to decrease computational costs, while algorithmic methods are often ignored. A quasistatic approach, which is based on calculating electric capacitance with the method of moments [1], is widely used to decrease computational costs of strip structure analysis in contrast to the electrodynamic approach. The method of moments implies solving linear systems with dense matrix. This paper presents the research into application of iterative methods for solving dense linear systems repeatedly during multivariant analysis of strip structures using the method of moments with the optimal choice of order of operations.

show abstract

Development of a CUDA Implementation of the 3D FDTD Method

Cited by 42 publications

References 7 publications

An FDTD-Based Parallel Virtual Tool for RCS Calculations of Complex Targets

An FDTD-Based Parallel Virtual Tool for RCS Calculations of Complex Targets

Efficient Parallel LOD-FDTD Method for Debye-Dispersive Media

Choosing order of operations to accelerate strip structure analysis in parameter range

Contact Info

Product

Resources

About