Supercomputer networking to accelerate large scale AI training

1 comments

• $mhandley 4 hours ago

  If you want more details, there's the protocol spec from OCP [0] and a paper [1]

  [0] https://opencompute.org/documents/ocp-mrc-1-0-pdf

  [1] https://cdn.openai.com/pdf/resilient-ai-supercomputer-networ...

  [I work on this]