NUMA-aware FFT-based Convolution on ARMv8 Many-core CPUs

TL;DR

This paper presents a NUMA-aware FFT-based convolution method optimized for ARMv8 many-core CPUs, significantly improving performance by reducing remote memory access through data reordering and parallelization.

Contribution

It introduces a novel NUMA-aware implementation of FFT-based convolution tailored for ARMv8 many-core CPUs, addressing non-uniform memory access challenges.

Findings

Significant performance gains over existing methods

Effective reduction of remote memory access

Enhanced parallelization efficiency

Abstract

Convolutional Neural Networks (CNNs), one of the most representative algorithms of deep learning, are widely used in various artificial intelligence applications. Convolution operations often take most of the computational overhead of CNNs. The FFT-based algorithm can improve the efficiency of convolution by reducing its algorithm complexity, there are a lot of works about the high-performance implementation of FFT-based convolution on many-core CPUs. However, there is no optimization for the non-uniform memory access (NUMA) characteristics in many-core CPUs. In this paper, we present a NUMA-aware FFT-based convolution implementation on ARMv8 many-core CPUs with NUMA architectures. The implementation can reduce a number of remote memory access through the data reordering of FFT transformations and the three-level parallelization of the complex matrix multiplication. The experiment results on a ARMv8 many-core CPU with NUMA architectures demonstrate that our NUMA-aware implementation has much better performance than the state-of-the-art work in most cases.