A Performance-Portable, Massively Parallel Distributed Nonuniform FFT

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fischill, Paul, Adelmann, Andreas, Muralikrishnan, Sriramkrishnan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916000997310464
author Fischill, Paul
Adelmann, Andreas
Muralikrishnan, Sriramkrishnan
author_facet Fischill, Paul
Adelmann, Andreas
Muralikrishnan, Sriramkrishnan
contents The nonuniform fast Fourier transform (NUFFT) enables spectral methods for problems with irregularly spaced samples, with applications in medical imaging, molecular dynamics, and kinetic plasma simulations. Existing implementations are limited to shared-memory execution, restricting problem sizes to what fits on a single node. We present the first distributed, performance-portable NUFFT for heterogeneous supercomputers. Our Kokkos-based implementation runs without modification on NVIDIA and AMD GPUs. We develop multiple spreading and interpolation kernels optimized for different accuracy requirements and architectures. Our spreading kernels match or exceed the single-GPU throughput of the state-of-the-art CUDA-based NUFFT library cuFINUFFT at production particle densities, while our Kokkos-based implementation additionally supports AMD GPUs. Strong scaling experiments on Alps (NVIDIA GH200), JUWELS Booster (NVIDIA A100), and LUMI (AMD MI250X) demonstrate scaling up to 1024 GPUs. At scale, the distributed FFT is a significant part of the total runtime, making higher NUFFT accuracy less expensive. We apply the method to massively parallel Particle-in-Fourier simulations of Landau damping with up to $1024^3$ Fourier modes and 8.6 billion particles on Alps, JUWELS, and LUMI, demonstrating that distributed NUFFTs enable kinetic plasma simulations at resolutions previously inaccessible to spectral particle methods.
format Preprint
id arxiv_https___arxiv_org_abs_2605_10678
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Performance-Portable, Massively Parallel Distributed Nonuniform FFT
Fischill, Paul
Adelmann, Andreas
Muralikrishnan, Sriramkrishnan
Computational Engineering, Finance, and Science
Mathematical Software
The nonuniform fast Fourier transform (NUFFT) enables spectral methods for problems with irregularly spaced samples, with applications in medical imaging, molecular dynamics, and kinetic plasma simulations. Existing implementations are limited to shared-memory execution, restricting problem sizes to what fits on a single node. We present the first distributed, performance-portable NUFFT for heterogeneous supercomputers. Our Kokkos-based implementation runs without modification on NVIDIA and AMD GPUs. We develop multiple spreading and interpolation kernels optimized for different accuracy requirements and architectures. Our spreading kernels match or exceed the single-GPU throughput of the state-of-the-art CUDA-based NUFFT library cuFINUFFT at production particle densities, while our Kokkos-based implementation additionally supports AMD GPUs. Strong scaling experiments on Alps (NVIDIA GH200), JUWELS Booster (NVIDIA A100), and LUMI (AMD MI250X) demonstrate scaling up to 1024 GPUs. At scale, the distributed FFT is a significant part of the total runtime, making higher NUFFT accuracy less expensive. We apply the method to massively parallel Particle-in-Fourier simulations of Landau damping with up to $1024^3$ Fourier modes and 8.6 billion particles on Alps, JUWELS, and LUMI, demonstrating that distributed NUFFTs enable kinetic plasma simulations at resolutions previously inaccessible to spectral particle methods.
title A Performance-Portable, Massively Parallel Distributed Nonuniform FFT
topic Computational Engineering, Finance, and Science
Mathematical Software
url https://arxiv.org/abs/2605.10678