1 Hit in 1.8 sec

cuFINUFFT: a load-balanced GPU library for general-purpose nonuniform FFTs [article]

Yu-hsuan Shih, Garrett Wright, Joakim andén, Johannes Blaschke, Alex H. Barnett
2021 arXiv   pre-print
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  ...  CONCLUSIONS We presented a general-purpose GPU-based library for nonuniform fast Fourier transforms: cuFINUFFT.  ... 
arXiv:2102.08463v2 fatcat:ttljkinggvfb5ogtug3bn2mk3e