Accelerating Finite Impulse Response Filtering Using Tensor Cores
Accelerating Finite Impulse Response Filtering Using Tensor Cores
复制标题
DOI：
--
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Takumi Kondo;Y. Maeda;Norishige Fukushima
中科院分区：
文献类型：
--
作者：
Takumi Kondo;Y. Maeda;Norishige Fukushima
本文研究了如何使用NVIDIA的张量核心加速单个通道2D图像卷积。 Tensor Core是一个专用的算术单元，用于加速矩阵产品，并提议加快机器学习中的卷积。从NVIDIA GPU的Volta架构中，张量核心用于各种应用。张量核心不仅限于机器学习，而且还限于各种应用。本文将有限冲动响应（FIR）过滤的单通道滤波器的图像卷积聚焦。通常，张量芯用于多通道卷积。因此，张量核的直接使用无法最大程度地提高常规FIR过滤的性能。因此，我们提出了三种用于FIR过滤的平行处理的编程模式：内核环展开，内核图像环的展开以及对角线内核图像环的展开。我们的实验比较了线性时间传播（LTI）和变体过滤器（LTV）的提议的环路展开方法：高斯滤波和双边滤波。结果表明，张量芯可以加速LTI过滤器的卷积，但无法加速线性LTV过滤器。另外，我们比较图灵和安培的两个GPU架构，并且可以在两个体系结构中使用TCS加速LTI滤波器。
This paper studies how to accelerate a single channel 2D image convolution using NVIDIA's Tensor Core. Tensor Core is a dedicated arithmetic unit for speeding up matrix products and was proposed to speed up the convolution in machine learning. From the Volta architecture of NVIDIA's GPU, Tensor Core is utilized for various applications. The Tensor Core is not limited to machine learning but various applications. This paper focus acceleration of image convolution of a single-channel filter of finite impulse response (FIR) filtering. Usually, Tensor Core is used for multi-channel convolution; thus, direct usage of Tensor Core cannot maximize the performance of the usual FIR filtering. Therefore, we propose three programming patterns for parallel processing of FIR filtering: kernel-loop unrolling, kernel-image-loop unrolling, and diagonal kernel-image-loop unrolling. Our experiments compare the proposed loop unrolling methods for linear time-invariant (LTI) and variant filters (LTV): Gaussian filtering and bilateral filtering. The results show that Tensor Core can accelerate the convolution for LTI filters but cannot accelerate linear LTV filters. Also, we compare the two GPU architectures of Turing and Ampere, and we can accelerate the LTI filter using TCs in both architectures.