~/nvidia-ncp-ari · AI Infrastructure Basics for Data Centers ▊
← NVIDIA-Certified Professional AI Rack and Interconnect
NVIDIA · objective · 20% of the exam
AI Infrastructure Basics for Data Centers — NVIDIA-Certified Professional AI Rack and Interconnect
The official NVIDIA documentation our AI Infrastructure Basics for Data Centers practice questions are cited to. Review the primary sources, then practise.
Official references for this objective
-
Nvidia — Data Centers Built for Advanced AI Reasoning | NVIDIA
NVLink and NVLink Switch High-speed interconnect for multi-GPU communication and large AI models
-
Nvidia — Networking — NVIDIA DGX SuperPOD: Data Center Design Featuring NVIDIA DGX H100 Systems
a tray with a cross sectional area of 11.5 in2 (7,419 mm2) is the minimum size that will support the volume of cables in an
-
Nvidia — Accelerated Computing Solutions | NVIDIA
NVIDIA Blackwell Ultra Delivers up to 50x Better Performance and 35x Lower Cost for Agentic AI
-
Nvidia — Magnum IO Software Stack for Accelerated Data Centers | NVIDIA
the GPU bypasses the CPU and system memory, and accesses remote storage via 8x 200 Gb/s NICs, achieving up to 1.6 TB/s of raw storage