AccFFT: A library for distributed-memory FFT on CPU and GPU architectures

TL;DR

AccFFT is a new distributed-memory FFT library optimized for CPU and GPU architectures, enabling scalable high-performance Fourier transforms across large GPU clusters.

Contribution

It extends existing FFT libraries to distributed GPU clusters using overlapping communication to reduce data transfer overhead.

Findings

Scales efficiently up to 4,096 GPUs on Titan

Reduces PCIe transfer overhead with overlapping communication

Demonstrates high performance on TACC and ORNL systems

Abstract

We present a new library for parallel distributed Fast Fourier Transforms (FFT). The importance of FFT in science and engineering and the advances in high performance computing necessitate further improvements. AccFFT extends existing FFT libraries for CUDA-enabled Graphics Processing Units (GPUs) to distributed memory clusters. We use overlapping communication method to reduce the overhead of PCIe transfers from/to GPU. We present numerical results on the Maverick platform at the Texas Advanced Computing Center (TACC) and on the Titan system at the Oak Ridge National Laboratory (ORNL). We present the scaling of the library up to 4,096 K20 GPUs of Titan.