PFFT: An extension of FFTW to massively parallel architectures

M Pippig - SIAM Journal on Scientific Computing, 2013 - SIAM
… We present an MPI based software library for computing fast Fourier transforms (FFTs) on …
Similar to established transpose FFT algorithms, we propose a parallel FFT framework that is …

Parallel three-dimensional nonequispaced fast Fourier transforms and their application to particle simulation

M Pippig, D Potts - SIAM Journal on Scientific Computing, 2013 - SIAM
PFFT is the only publicly available parallel FFT software library that pays special attention to
the efficient parallel … ˆRk of the regularized kernel function and completed by a parallel, three-…

[PDF][PDF] PFFT User Manual

M Pippig - 2018 - www-user.tu-chemnitz.de
… We explain the basic steps for computing a parallel FFT with the PFFT library at the example
of … of a single three-dimensional parallel FFT of size n[0] x n[1] x n[2] is done by the function

AccFFT: A library for distributed-memory FFT on CPU and GPU architectures

A Gholami, J Hill, D Malhotra, G Biros - arXiv preprint arXiv:1506.07933, 2015 - arxiv.org
… a new library for parallel distributed Fast Fourier Transforms (FFT). The importance of FFT
in … PFFT has an auto-tuning function for finding an optimal communication pattern. A very …

An efficient and flexible parallel FFT implementation based on FFTW

M Pippig - Competence in High Performance Computing 2010 …, 2011 - Springer
library called PFFT [12], which was developed for calculating … , we compare features of
P3DFFT and our new PFFT … If we call this library function with the flags …

Performance optimization of multithreaded 2d fast fourier transform on multicore processors using load imbalancing parallel computing method

S Khokhriakov, RR Manumachu, A Lastovetsky - IEEE Access, 2018 - ieeexplore.ieee.org
… provide libraries containing 1D, 2D, and 3D FFT routines … two optimization methods, PFFT-FPM
and PFFT-FPM-PAD … The discrete speed function of processor Pi is given by Si = {si(x1,y1…

Fast parallel multidimensional FFT using advanced MPI

L Dalcin, M Mortensen, DE Keyes - … of Parallel and Distributed Computing, 2019 - Elsevier
… and GPUs, the parallel FFT subroutine library of Plimpton [28]… PSPFFT, which provides a
parallel FFT-based Poisson solver … present in PFFT and P3DFFT, these libraries have options to …

tcfft: A fast half-precision fft library for nvidia tensor cores

B Li, S Cheng, J Lin - 2021 IEEE International Conference on …, 2021 - ieeexplore.ieee.org
… 16×16 matrices which can be calculated in parallel and 16×16 matrices can fill Tensor Core
… For measuring time, we only count the time spent on execute function; the time spent on …

FluidFFT: Common API (C++ and Python) for fast fourier transform HPC libraries

AV Mohanan, C Bonamy, P Augier - arXiv preprint arXiv:1807.01775, 2018 - arxiv.org
… (FFT) in sequential, in parallel and on GPU with different FFT libraries (FFTW, P3DFFT, PFFT,
… A helper function fluidfft.import fft class is provided with the package to simply import the …

FFT, FMM, or multigrid? A comparative study of state-of-the-art Poisson solvers for uniform and nonuniform grids in the unit cube

A Gholami, D Malhotra, H Sundar, G Biros - SIAM Journal on Scientific …, 2016 - SIAM
… P3DFFT is a robust parallel FFT library that has been successfully … The second library is
PFFT which has been recently … function f with K(x) = −(4π|x|)−1 (free space Green’s function for …