• Corpus ID: 5822392

Compute unified device architecture (CUDA) based finite-difference time-domain (FDTD) implementation

@article{Demir2010ComputeUD,
  title={Compute unified device architecture (CUDA) based finite-difference time-domain (FDTD) implementation},
  author={Veysel Demir and Atef Z. Elsherbeni},
  journal={Applied Computational Electromagnetics Society Journal},
  year={2010},
  volume={25},
  pages={303-314}
}
  • V. Demir, A. Elsherbeni
  • Published 1 April 2010
  • Computer Science
  • Applied Computational Electromagnetics Society Journal
Recent developments in the design of graphics processing units (GPUs) have made it possible to use these devices as alternatives to central processor units (CPUs) and perform high performance scientific computing on them. Though several implementations of finite- difference time-domain (FDTD) method have been reported, the unavailability of high level languages to program graphics cards had been a major obstacle for scientists and engineers who would want to develop codes for graphics cards… 
CUDA Fortran acceleration for the finite-difference time-domain method
TLDR
The derived CUDA Fortran code is compared with an optimized CPU version that runs on a workstation-class CPU to present a realistic GPU to CPU run time comparison and thus help in making better informed investment decisions on FDTD code redesigns and equipment upgrades.
Improved performance of FDTD computation using a thread block constructed as a two-dimensional array with CUDA
TLDR
The authors investigated the computational performance with respect to the size of a thread block constructed as a 2-D array, and improved the performance of the implementation.
Performance Optimization of Massively Parallel FDTD Computations
TLDR
The NVIDIA’s Compute Unified Device Architecture CUDA on an NVIDIA GeForce processor will be used to solve a Finite Difference Time Domain FDTD computation, with the focus on optimizing the efficiency of the algorithm by maximizing the throughput through affective use of the fast on-chip shared memory, and avoid using the slow off-chip global memory.
Graphics processor unit (GPU) acceleration of finite-difference frequency-domain (FDFD) method
TLDR
It is shown that FDFD can be solved with a speed-up factor of more than 20 on a GPU compared with the solution on a central processing unit (CPU), while memory usage can be reduced substantially with the presented algorithm.
A comparison of computing architectures and parallelization frameworks based on a two-dimensional FDTD
TLDR
Assessment and compares the performance of different parallelization frameworks and different multicore architectures for exploiting parallelism when Maxwell's Equations have to be solved and results show that the recent GPU architecture, Kepler, did not outperform the older Fermi architecture.
FDTD Acceleration using MATLAB Parallel Computing Toolbox and GPU
TLDR
A MATLAB based finite difference time domain (FDTD) method accelerated using the GPU functions in MATLAB’s parallel computing toolbox (PCT) is presented and several modifications to increase the efficiency on several different NVIDIA graphics cards are demonstrated.
FDTD on Distributed Heterogeneous Multi-GPU Systems
TLDR
A new FDTD decomposition implementation that can be executed on a cluster of heterogeneous systems with a multi-core CPU, and one or several CUDA capable GPUs, and a performance increase of 66% when simulating large domains on two GPUs compared to a single GPU.
GPU‐based acceleration of computational electromagnetics codes
TLDR
An overview of the main efforts of researchers to port computational electromagnetics codes to GPU is given and GPU implementation aspects of two well-known techniques, namely the finite-difference time domain (FDTD) and the method of moments (MoM), are investigated.
GPU accelerated finite-element computation for electromagnetic analysis
TLDR
This paper identifies the bottlenecks in the GPU parallelization of the Finite-Element Method for electromagnetic analysis, and proposes potential solutions to alleviate the bottlenecking, and shows that with a proper parallelization and implementation, GPUs are able to achieve significant speedups over OpenMP-enabled multi-core CPUs.
Acceleration of the Dual-Field Domain Decomposition Algorithm Using MPI–CUDA on Large-Scale Computing Systems
TLDR
This paper is able to achieve a significant speedup by utilizing GPUs in a message-passing interface (MPI)-based cluster environment and the same acceleration strategy can be applied to the acceleration of the discontinuous Galerkin time-domain (DGTD) algorithms.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 29 REFERENCES
Graphics processor unit (GPU) acceleration of finite-difference time-domain (FDTD) algorithm
TLDR
Off-the-shelf graphics processor units (GPUs) can be successfully used to accelerate FDTD simulations and it is demonstrated that the GPU outperforms a central processing unit (CPU) of comparable technology generation.
Finite Difference Time Domain (FDTD) Simulations Using Graphics Processors
  • S. Adams, J. Payne, R. Boppana
  • Computer Science
    2007 DoD High Performance Computing Modernization Program Users Group Conference
  • 2007
TLDR
This paper presents a graphics processor based implementation of the Finite Difference Time Domain, which uses a central finite differencing scheme for solving Maxwell's equations for electromagnetics and shows how GPUs can be used to greatly speedup FDTD simulations.
Acceleration of finite-difference time-domain (FDTD) using graphics processor units (GPU)
TLDR
It is demonstrated that standard consumer Graphics Processor Units (GPUs) can be used to accelerate FDTD simulations by a factor of over seven, relative to an Intel CPU of similar technology generation.
A GPU implementation of the 2-D finite-difference time-domain code using high level shader language
TLDR
A graphics processing unit (GPU) is applied to the finite-difference timedomain (FDTD) method to realize a cost-effective and high-speed computation of an FDTD simulation and achieves an approximately 10-fold improvement in computation speed.
FDTD calculations using graphical processing units
  • M. J. Inman, A. Elsherbeni, C.E. Smith
  • Computer Science
    IEEE/ACES International Conference on Wireless Communications and Applied Computational Electromagnetics, 2005.
  • 2005
TLDR
This paper deals with using tools commonly available to programmers to implement the finite difference time domain (FDTD) calculations using video cards to maximize the processing power available to an extent greater than possible without the addition of the video card.
GPU-based accelerated 2D and 3D FDTD solvers
TLDR
This paper presents GPU-based accelerated solvers for the Finite-Difference Time-Domain method in both its 2D and 3D embodiments.
High-speed FDTD simulation algorithm for GPU with compute unified device architecture
TLDR
The GPU-FDTD algorithm performed high-speed FDTD simulation using GPU with CUDA, and maintained single-floating point accuracy, and suggests that the bottleneck of theFDTD simulation is memory bandwidth.
Accelerating Simulations of Light Scattering Based on Finite-Difference Time-Domain Method with General Purpose GPUs
TLDR
This paper examines how the computation time of the finite-difference time-domain (FDTD), a classic numerical method for computing a solution to Maxwell's equations, can be reduced by leveraging the massively parallel architecture of GPGPUs cards.
Programming video cards for computational electromagnetics applications
TLDR
By formulating proper procedures to realize general vector computations on the GPU, it will be possible to increase the processing power available by at least an order of magnitude compared to the current generation of CPUs.
Optimization and parameter exploration using GPU based FDTD solvers
TLDR
This paper will detail the use of a developed GPU based FDTD solver for both optimization and through a graphical interface parameter exploration.
...
1
2
3
...