DGX
NVIDIA's fully integrated AI server and SuperPOD reference system sold as a turnkey appliance.
Current numbers
0.5–5 GB/s/GPUDGX H200 SuperPOD shared-storage read guidelines: 0.5, 1, and 5 GB/s per GPU for Good, Better, and Best performance, with 4 GB/s/GPU recommended for computer-vision workloads
250 / 125 GB/sper scalable unit read / write — DGX B300 SuperPOD 'Enhanced' storage tier
250 / 125 GB/sper scalable unit, DGX B300 SuperPOD 'Enhanced' storage tier (read/write); 8-SU = 2,000 / 1,000 GB/s