Demystifying GPU Compute Architectures - by Babbage

Getting 'low level' with Nvidia and AMD GPUs

NVIDIA_GeForce_8800GTX_G80___DSC01200 via Fritzchens Fritz https://commons.wikimedia.org/w/index.php?curid=65926359 The complexity of modern GPUs needn’t obscure the essence of how these machines work. If we are to make the most of the advanced technology in these GPUs then we need a little less mystique. I think this is one of the most exciting times in computer architecture for many years. In CPUs we have x86, Arm and RISC-V, three increasingly well-supported application architectures. And GPU architectures are, of course, becoming much more important. GPUs are obviously a critical…

related reading

• Demystifying GPU Compute Architectures - by Babbagethechipletter.substack.com
• AI Chip Architecturesjacobpeake.com
• Inside NVIDIA GPUs: Anatomy of high performance matmul kernels - Aleksa Gordićaleksagordic.com
• How to Think About GPUs | How To Scale Your Modeljax-ml.github.io
• The Best GPUs for Deep Learning in 2023 — An In-depth Analysistimdettmers.com
• A history of NVidia Stream Multiprocessorfabiensanglard.net
• GPU Performance Background User's Guide - NVIDIA Docsdocs.nvidia.com
• AI Chip Architecturesjepeake.com
• How to Optimize a CUDA Matmul Kernel for cuBLAS-like Performance: a Worklogsiboehm.com
• Chip design from the bottom up – Reiner Popeyoutube.com
• The Best GPUs for Deep Learning in 2023 — An In-depth Analysistimdettmers.com
• GitHub - adam-maj/tiny-gpu: A minimal GPU design in Verilog to learn how GPUs work from the ground up · GitHubgithub.com