NVIDIA A100 Tensor Core GPU: Performance and Innovation

被引：216

作者：

Choquette, Jack ^{[1
]}

Gandhi, Wishwesh ^{[2
]}

Giroux, Olivier ^{[1
]}

Stam, Nick ^{[1
]}

Krashinsky, Ronny ^{[1
]}

机构：

[1] NVIDIA, Singapore 609927, Singapore

[2] NVIDIA, Architecture, Singapore 609927, Singapore

来源：

IEEE MICRO | 2021年 / 41卷 / 02期

关键词：

A100; C++20; CUDA; Deep Learning; GPU; NVLink; Tensor Core;

D O I：

10.1109/MM.2021.3061394

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

NVIDIA A100 Tensor Core GPU is NVIDIA's latest flagship GPU. It has been designed with many new innovative features to provide performance and capabilities for HPC, AI, and data analytics workloads. Feature enhancements include a Third-Generation Tensor Core, new asynchronous data movement and programming model, enhanced L2 cache, HBM2 DRAM, and third-generation NVIDIA NVLink I/O. © 1981-2012 IEEE.

引用

页码：29 / 35

页数：7