Opening Remarks
summary
Opening remarks set the agenda: AI networking as an end-to-end problem spanning data center, backbone and edge; operations must scale as model training grows from Llama 2 on a couple thousand GPUs to Llama 3 on 16,000 GPUs and 24,000‑GPU clusters, with GPUs CPUs memory NICs optics and switches all tracked and repaired fast because training jobs take weeks or months. Panels follow each session and attendees can submit questions via QR codes for live discussion.
AI-written summary · about
