AI & machine learning2023 · shared on Hacker News
TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning
Explore TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning, its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.
GO BEYOND THE ABSTRACT
The conversation around the idea.
Original Hacker News threads. Different perspectives, questions, and things worth thinking about.
Showing up to 100 related threads. A paper may appear in a comment rather than the original submission.
Links found in the discussion
Up to 50 recent source links, in HN item order.
Ytranspute ·
TPU Deep Dive
451 points86 comments
Ymfiguiere ·
How Meta trains large language models at scale
404 points194 comments
Yibobev ·
Basic Facts about GPUs
344 points81 comments
Ywalterbell ·
OpenTPU: Open-Source Reimplementation of Google Tensor Processing Unit (TPU)
166 points22 comments
Ymariuz ·
TPU transformation: A look back at 10 years of our AI-specialized chips
142 points62 comments
Yxtoilette ·
IP over Lasers
121 points80 comments
Yzekrioca ·
TPU v4 provides exaFLOPS-scale ML with industry-leading efficiency
5 points0 comments
YMacsHeadroom ·
Nvidia DGX GH200: The First 100 Terabyte GPU Memory System
5 points1 comments
Ymfiguiere ·
TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning
1 points0 comments
Yzdw ·
HPC is dying, and MPI is killing it
1 points0 comments
Yjonbaer ·
Google Launches AI Supercomputer Powered by Nvidia H100 GPUs
1 points0 comments
Ythinxer ·
MTIA v1: Meta’s first-generation AI inference accelerator
1 points0 comments