kokkos-fft: A shared-memory FFT for the Kokkos ecosystem
Open Access
- 30 July 2025
- journal article
- Published by The Open Journal in Journal of Open Source Software
- Vol. 10 (111) , 8391
Abstract
Kokkos-fft provides a unified, performance-portable interface for Fast Fourier Transforms (FFTs) within the Kokkos ecosystem (C. Trott et al., 2021). It seamlessly integrates with leading local FFT libraries including FFTW, cuFFT, rocFFT, and oneMKL. Designed for simplicity and efficiency, kokkos-fft offers a user experience akin to numpy.fft for in-place and out-of-place transforms, while leveraging the raw speed of vendor-optimized libraries. A demonstration solving 2D Hasegawa-Wakatani turbulence with the Fourier spectral method illustrates how kokkos-fft can deliver significant speedups over Python-based alternatives without drastically increasing code complexity, empowering researchers to perform high-performance FFTs simply and effectively.Keywords
This publication has 6 references indexed in Scilit:
- The Kokkos EcoSystem: Comprehensive Performance Portability for High Performance ComputingComputing in Science & Engineering, 2021
- Kokkos 3: Programming Model Extensions for the Exascale EraIEEE Transactions on Parallel and Distributed Systems, 2021
- Array programming with NumPyNature, 2020
- FluidFFT: Common API (C++ and Python) for Fast Fourier Transform HPC LibrariesJournal of Open Research Software, 2019
- The FFT: an algorithm the whole family can useComputing in Science & Engineering, 2000
- A collisional drift wave description of plasma edge turbulenceThe Physics of Fluids, 1984