NVIDIA Investor Presentation

Full-Stack & Data Center Scale Acceleration Drive Significant Cost Savings and Workload Scaling Classical Computing - 92 CPU-only servers $3.3M (including, switches, cables, racks) Application CPU server racks Cost comparison example based on latest available NVIDIA A100 GPU and Intel CPU inference results in the commercially available category of the MLPerf industry benchmark; includes related infrastructure costs such as networking. Accelerated Computing - 1 NVIDIA DGX A100 $220,000 DGX and $100,000 NVIDIA AI software Application Re-Engineered for Acceleration CUDA-X Acceleration Libraries Magnum 10 DGX 10X lower cost 14X better energy-efficiency NVIDIA.

View entire presentation