Multilevel fast multipole algorithm enhanced by GPU parallel technique for electromagnetic scattering problems

Xu, Kun; Ding, D. Z.; Zheng, Fan; Chen, R. S.

doi:10.1002/mop.24963

Cited by 38 publications

(13 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These two steps must be balanced by changing the level of box divisions according to the number of particles being calculated. The related discussion can be found in [18][19][20].…”

Section: Fast Multipole Methodsmentioning

confidence: 99%

Acceleration of Vortex Methods Calculation Using FMM and Mdgrape-3

Sheel¹

2011

PIER B

View full text Add to dashboard Cite

Abstract-The present study discusses some numerical techniques on the simultaneous use of the Fast Multipole Method (FMM) and special-purpose computer (MDGRAPE-3) to make the impractically expensive calculation feasible without the loss of numerical accuracy. In the present calculations, the impingement of two identical inclined vortex rings has been studied, and the computation time has been reduced by a factor of 1000 at N = 1.18×10 6 where N is the number of vortex elements. The direct and MDGRAPE-3 calculations both have a scaling of O(N 2 ), and the use of the FMM brings them both down to O(N ). The global kinetic energy, enstrophy and energy spectra have been investigated to address the numerical accuracy and have good agreement with other similar works.

show abstract

“…These two steps must be balanced by changing the level of box divisions according to the number of particles being calculated. The related discussion can be found in [18][19][20].…”

Section: Fast Multipole Methodsmentioning

confidence: 99%

Acceleration of Vortex Methods Calculation Using FMM and Mdgrape-3

Sheel¹

2011

PIER B

View full text Add to dashboard Cite

show abstract

“…For example, fixed indexes for loops determine the iterations. The loop iterations must be countable (6) The loop body cannot contain any function calls other than the standard internal function On the application of VALU acceleration to the MLFMA code, some important issues should be addressed. First, only the calculations of the same type are concurrently computed by VALU in order to achieve the best vectorization performance of VALU.…”

Section: Valu Hardware Acceleration Techniquementioning

confidence: 99%

Parallelization of the multilevel fast multipole algorithm using the OpenMP and VALU acceleration

Wang

2013

2013 IEEE INTERNATIONAL CONFERENCE ON MICROWAVE TECHNOLOGY &Amp; COMPUTATIONAL ELECTROMAGNETICS

View full text Add to dashboard Cite

In this paper, a parallelization scheme by the combined use of the OpenMP and Vector Arithmetic Logic Unit (VALU) hardware acceleration is proposed to speed up the multilevel fast multipole algorithm (MLFMA) on the multicore computer architectures. This accelerating technology doesn't need to add any additional hardware equipment and could be applied to any other multicore computers whose all can obtain good accelerating effect. In addition, the spherical harmonics expansion based MLFMA is applied for reducing the memory costed. Some numerical results of typical objects are shown, through which comparative analyses are performed to give some concluding remarks.

show abstract

“…The programming combination between GPU and CPU can fully depend on the easy-to-use language CUDA-C/PTX. To the authors' knowledge, the GPU/CPU heterogeneous platform is the most popular choice especially in CEM, such as the GPU-based FDTD [15], MLFMM [16][17][18][19], AIM [20], P-FFT [21], MoM [22,23], and higher-order MoM [24]. In 2013, an impressive implementation of MLFMM by OpenMP-CUDA was realized on Fermi architecture (NVIDIA Tesla C2050), which achieved much higher performance than those before [16].…”

Section: Introductionmentioning

confidence: 99%

High Performance Computing of Complex Electromagnetic Algorithms Based on GPU/CPU Heterogeneous Platform and Its Applications to EM Scattering and Multilayered Medium Structure

Song

Mu²,

Zhou

2017

International Journal of Antennas and Propagation

View full text Add to dashboard Cite

The fast and accurate numerical analysis for large-scale objects and complex structures is essential to electromagnetic simulation and design. Comparing to the exploration in EM algorithms from mathematical point of view, the computer programming realization is coordinately significant while keeping up with the development of hardware architectures. Unlike the previous parallel algorithms or those implemented by means of parallel programming on multicore CPU with OpenMP or on a cluster of computers with MPI, the new type of large-scale parallel processor based on graphics processing unit (GPU) has shown impressive ability in various scenarios of supercomputing, while its application in computational electromagnetics is especially expected. This paper introduces our recent work on high performance computing based on GPU/CPU heterogeneous platform and its application to EM scattering problems and planar multilayered medium structure, including a novel realization of OpenMP-CUDA-MLFMM, a developed ACA method and a deeply optimized CG-FFT method. With fruitful numerical examples and their obvious enhancement in efficiencies, it is convincing to keep on deeply investigating and understanding the computer hardware and their operating mechanism in the future.

show abstract

Multilevel fast multipole algorithm enhanced by GPU parallel technique for electromagnetic scattering problems

Cited by 38 publications

References 13 publications

Acceleration of Vortex Methods Calculation Using FMM and Mdgrape-3

Acceleration of Vortex Methods Calculation Using FMM and Mdgrape-3

Parallelization of the multilevel fast multipole algorithm using the OpenMP and VALU acceleration

High Performance Computing of Complex Electromagnetic Algorithms Based on GPU/CPU Heterogeneous Platform and Its Applications to EM Scattering and Multilayered Medium Structure

Contact Info

Product

Resources

About