By Yu-hsuan Shih, Garrett Wright, Joakim Andén, Johannes Blaschke, Alex H. Barnett

Nonuniform fast Fourier transforms dominate the computational cost in many applications including image reconstruction and signal processing. We thus present a general-purpose GPU-based CUDA library for type 1 (nonuniform to uniform) and type 2 (uniform to nonuniform) transforms in dimensions 2 and 3, in single or double precision. It achieves high performance for a given user-requested accuracy, regardless of the distribution of nonuniform points, via cache-aware point reordering, and load-balanced blocked spreading in shared memory. At low accuracies, this gives on-GPU throughputs around 10e9 nonuniform points per second, and (even including host-device transfer) is typically 4-10x faster than the latest parallel CPU code FINUFFT (at 28 threads). It is competitive with two established GPU codes, being up to 90x faster at high accuracy and/or type 1 clustered point distributions. Finally we demonstrate a 5-12x speedup versus CPU in an X-ray diffraction 3D iterative reconstruction task at 10e-12 accuracy, observing excellent multi-GPU weak scaling up to one rank per GPU.

Read the paper: https://doi.org/10.48550/arXiv.2102.08463