arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.18328cs.AR
收藏

开源 GEMM 硬件内核生成器:迈向数值定制化计算

Open-Source GEMM Hardware Kernels Generator: Toward Numerically-Tailored Computations

Louis Ledoux, Marc Casas

AI总结:

本文提出一种开源 GEMM 硬件内核生成器,旨在针对从高精度科学计算到低精度深度神经网络等不同数值需求,实现数值定制化的矩阵乘法计算。

摘要:

许多科学计算问题可以归结为矩阵-矩阵乘法(MMM),这使得基础线性代数子程序(BLAS)中的通用矩阵乘法(GEMM)内核受到高性能计算社区的关注。然而,这些工作负载具有广泛的数值需求。病态线性系统需要高精度算术以确保结果正确且可复现。相比之下,新兴工作负载(如深度神经网络)可能拥有数百万到数十亿个参数,已表现出对算术调整和精度降低的鲁棒性。

英文摘要:

Many scientific computing problems can be reduced to Matrix-Matrix Multiplications (MMM), making the General Matrix Multiply (GEMM) kernels in the Basic Linear Algebra Subroutine (BLAS) of interest to the high-performance computing community. However, these workloads have a wide range of numerical requirements. Ill-conditioned linear systems require high-precision arithmetic to ensure correct and reproducible results. In contrast, emerging workloads such as deep neural networks, which can have millions up to billions of parameters, have shown resilience to arithmetic tinkering and precision lowering.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑