High Scalability FMM-FFT Electromagnetic Solver for Supercomputer Systems

Taboada, J. M.; Landesa, L.; Obelleiro, F.; Rodríguez, J.L.; Bertolo, J. M.; Araújo, M. G.; Mouriño, J. Carlos; Gómez, A.

doi:10.1109/map.2009.5433091

Cited by 50 publications

(31 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The difficulties faced in successfully parallelizing it in computer clusters have been widely discussed by many research groups [8,10,14,15]. Considering the favorable MLFMA features in shared-memory computers, and taking into account the high scalability behavior of the FMM-FFT in distributed computers while maintaining a low numerical complexity, a proper combination of both techniques seems to be a good alternative to take a step further after the works of [19] and [22]. The MLFMA-FFT presented here combines both methods to get an optimal scheme from the point of view of modern mixed-memory supercomputers.…”

Section: The Proposed Method: Parallel Mlfma-fftmentioning

confidence: 99%

“…Additionally, the use of the FFT to accelerate the translation stage provides a great reduction of the MVP CPU time with respect to the conventional FMM. An electromagnetic problem of more than 150 million unknowns has been solved in [19] using a hybrid parallel implementation of the method. We have used the MPI library for communications between distributed nodes and OpenMP standard for threads inside each shared-memory node.…”

Section: Background: Parallel Fmm-fft and Nested Fmm-fftmentioning

confidence: 99%

“…A distribution by unknowns is used for the parallel iterative solver. Further details of this FMM-FFT implementation can be found in [19].…”

Section: Background: Parallel Fmm-fft and Nested Fmm-fftmentioning

confidence: 99%

“…The FMM-FFT preserves the natural parallel scaling propensity of the single-level FMM in the spectral (k-space) domain [18,19]. Moreover, it requires less time than the FMM for computing the matrix-vector product (MVP) due to the translation stage acceleration given by the FFT.…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, it requires less time than the FMM for computing the matrix-vector product (MVP) due to the translation stage acceleration given by the FFT. This method has demonstrated to be a good alternative to benefit from massively parallel distributed computers [19]. Recently, a hybrid Message Passing Interface (MPI)/OpenMP parallel implementation of a nested scheme of the FMM-FFT was proposed by the authors in [20][21][22].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Mlfma-FFT Parallel Algorithm for the Solution of Large-Scale Problems in Electromagnetics

Taboada¹,

Araújo²,

Bertolo³

et al. 2010

PIER

100

View full text Add to dashboard Cite

Abstract-An efficient hybrid MPI/OpenMP parallel implementation of an innovative approach that combines the Fast Fourier Transform (FFT) and Multilevel Fast Multipole Algorithm (MLFMA) has been successfully used to solve an electromagnetic problem involving 620 millions of unknowns. The MLFMA-FFT method can deal with extremely large problems due to its high scalability and its reduced computational complexity. The former is provided by the use of the FFT in distributed calculations and the latter by the application of the MLFMA in shared computation.

show abstract

Section: The Proposed Method: Parallel Mlfma-fftmentioning

confidence: 99%

Section: Background: Parallel Fmm-fft and Nested Fmm-fftmentioning

confidence: 99%

“…A distribution by unknowns is used for the parallel iterative solver. Further details of this FMM-FFT implementation can be found in [19].…”

Section: Background: Parallel Fmm-fft and Nested Fmm-fftmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Mlfma-FFT Parallel Algorithm for the Solution of Large-Scale Problems in Electromagnetics

Taboada¹,

Araújo²,

Bertolo³

et al. 2010

PIER

100

View full text Add to dashboard Cite

show abstract

Radar Cross Section

Jenn

2024

Encyclopedia of RF and Microwave Engineering

View full text Add to dashboard Cite

Radar cross section (RCS) is a measure of the magnitude of the scattered electromagnetic (EM) wave from a body relative to the magnitude of the wave incident on the body. It is an indicator of the strength of an object's signal returned to a radar sensor. To many people, RCS is synonymous with stealth. Stealth technology has become an integral part of all military systems since it was first employed in the 1970s. However, RCS is just one aspect of stealth, the design philosophy that seeks to make a platform (such as an aircraft, ship, or ground vehicle) undetectable to a wide range of sensors. Recently, there has been a rapid expansion of radar sensors for commercial applications like autonomous vehicles and robotics. These applications necessitate the characterization of the RCS of a wide range of objects commonly found in the urban environment, such as cars, bicycles, and even pedestrians.

show abstract

References

2014

The Multilevel Fast Multipole Algorithm (MLFMA) for Solving Large‐Scale Computational Electromagnetics Problems

View full text Add to dashboard Cite

High Scalability FMM-FFT Electromagnetic Solver for Supercomputer Systems

Cited by 50 publications

References 21 publications

Mlfma-FFT Parallel Algorithm for the Solution of Large-Scale Problems in Electromagnetics

Mlfma-FFT Parallel Algorithm for the Solution of Large-Scale Problems in Electromagnetics

Radar Cross Section

References

Contact Info

Product

Resources

About