-
cuFINUFFT: a load-balanced GPU library for general-purpose nonuniform FFTs
Authors:
Yu-hsuan Shih,
Garrett Wright,
Joakim Andén,
Johannes Blaschke,
Alex H. Barnett
Abstract:
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,…
▽ More
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 $10^9$ nonuniform points per second, and (even including host-device transfer) is typically 4-10$\times$ faster than the latest parallel CPU code FINUFFT (at 28 threads). It is competitive with two established GPU codes, being up to 90$\times$ faster at high accuracy and/or type 1 clustered point distributions. Finally we demonstrate a 5-12$\times$ speedup versus CPU in an X-ray diffraction 3D iterative reconstruction task at $10^{-12}$ accuracy, observing excellent multi-GPU weak scaling up to one rank per GPU.
△ Less
Submitted 25 March, 2021; v1 submitted 16 February, 2021;
originally announced February 2021.
-
A parallel non-uniform fast Fourier transform library based on an "exponential of semicircle" kernel
Authors:
Alex H. Barnett,
Jeremy F. Magland,
Ludvig af Klinteberg
Abstract:
The nonuniform fast Fourier transform (NUFFT) generalizes the FFT to off-grid data. Its many applications include image reconstruction, data analysis, and the numerical solution of differential equations. We present FINUFFT, an efficient parallel library for type 1 (nonuiform to uniform), type 2 (uniform to nonuniform), or type 3 (nonuniform to nonuniform) transforms, in dimensions 1, 2, or 3. It…
▽ More
The nonuniform fast Fourier transform (NUFFT) generalizes the FFT to off-grid data. Its many applications include image reconstruction, data analysis, and the numerical solution of differential equations. We present FINUFFT, an efficient parallel library for type 1 (nonuiform to uniform), type 2 (uniform to nonuniform), or type 3 (nonuniform to nonuniform) transforms, in dimensions 1, 2, or 3. It uses minimal RAM, requires no precomputation or plan steps, and has a simple interface to several languages. We perform the expensive spreading/interpolation between nonuniform points and the fine grid via a simple new kernel---the `exponential of semicircle' $e^{β\sqrt{1-x^2}}$ in $x\in[-1,1]$---in a cache-aware load-balanced multithreaded implementation. The deconvolution step requires the Fourier transform of the kernel, for which we propose efficient numerical quadrature. For types 1 and 2, rigorous error bounds asymptotic in the kernel width approach the fastest known exponential rate, namely that of the Kaiser--Bessel kernel. We benchmark against several popular CPU-based libraries, showing favorable speed and memory footprint, especially in three dimensions when high accuracy and/or clustered point distributions are desired.
△ Less
Submitted 8 April, 2019; v1 submitted 20 August, 2018;
originally announced August 2018.
-
Validation of neural spike sorting algorithms without ground-truth information
Authors:
Alex H. Barnett,
Jeremy F. Magland,
Leslie F. Greengard
Abstract:
We describe a suite of validation metrics that assess the credibility of a given automatic spike sorting algorithm applied to a given electrophysiological recording, when ground-truth is unavailable. By rerunning the spike sorter two or more times, the metrics measure stability under various perturbations consistent with variations in the data itself, making no assumptions about the noise model, n…
▽ More
We describe a suite of validation metrics that assess the credibility of a given automatic spike sorting algorithm applied to a given electrophysiological recording, when ground-truth is unavailable. By rerunning the spike sorter two or more times, the metrics measure stability under various perturbations consistent with variations in the data itself, making no assumptions about the noise model, nor about the internal workings of the sorting algorithm. Such stability is a prerequisite for reproducibility of results. We illustrate the metrics on standard sorting algorithms for both in vivo and ex vivo recordings. We believe that such metrics could reduce the significant human labor currently spent on validation, and should form an essential part of large-scale automated spike sorting and systematic benchmarking of algorithms.
△ Less
Submitted 27 August, 2015;
originally announced August 2015.
