Year: 2024 · Venue: ACM SIGCOMM · Authors: Adithya Gangidi, Rui Miao, Shengbao Zheng, et al.
Find the paper on Google Scholar
Abstract
Operating RoCE fabrics for AI clusters with tens of thousands of GPUs. The 1988 congestion problem returns at 400 Gbps — and the story is far from over.
My Notes
I haven’t written up my notes for this one yet — this page is a placeholder that I’ll fill in as I read.