Fast parallel multidimensional FFT using advanced MPI

TL;DR

This paper introduces a novel MPI-based method for multidimensional FFTs that reduces local data realignment by using discontiguous memory communication, achieving performance comparable or superior to existing libraries.

Contribution

The paper presents a new MPI-2 based approach utilizing subarray datatypes and generalized all-to-all scatter/gather for efficient multidimensional array redistribution in parallel FFTs.

Findings

Performance on par or better than MPI-FFTW, P3DFFT, and 2DECOMP&FFT

Reduces local data realignments in multidimensional FFTs

Applicable to arbitrary array decompositions and processor grids

Abstract

We present a new method for performing global redistributions of multidimensional arrays essential to parallel fast Fourier (or similar) transforms. Traditional methods use standard all-to-all collective communication of contiguous memory buffers, thus necessary requiring local data realignment steps intermixed in-between redistribution and transform steps. Instead, our method takes advantage of subarray datatypes and generalized all-to-all scatter/gather from the MPI-2 standard to communicate discontiguous memory buffers, effectively eliminating the need for local data realignments. Despite generalized all-to-all communication of discontiguous data being generally slower, our proposal economizes in local work. For a range of strong and weak scaling tests, we found the overall performance of our method to be on par and often better than well-established libraries like MPI-FFTW, P3DFFT, and 2DECOMP&FFT. We provide compact routines implemented at the highest possible level using the MPI bindings for the C programming language. These routines apply to any global redistribution, over any two directions of a multidimensional array, decomposed on arbitrary Cartesian processor grids (1D slabs, 2D pencils, or even higher-dimensional decompositions). The high level implementation makes the code easy to read, maintain, and eventually extend. Our approach enables for future speedups from optimizations in the internal datatype handling engines within MPI implementations.