Scale-out
Connecting many nodes into a looser cluster fabric (InfiniBand or Ethernet) to scale beyond one coherent domain.
Current numbers
~18x (GB200, 400G) / ~9x (GB300, 800G) / ~7.5x (Vera Rubin, 1.6T)scale-up versus back-end scale-out peak bandwidth per GPU: roughly 8-18x per direction by generation
1.1, released 2026-10-12UEC scale-out specification; contract the implemented profile
~18× GB200 / ~9× GB300NVLink 5 per-GPU scale-up (~900 GB/s per direction) vs one scale-out NIC per GPU: ~18× GB200/CX-7/400G, ~9× GB300/CX-8/800G