← Hardware, Cabling & Cluster Build-Out / 22

hardNVIDIAMicrosoftCoreWeave

For 64 to 72 GPUs, is one GB300 NVL72 rack better than eight HGX B300 nodes?

Compare 64 or 72 useful GPUs, identify the communication that could improve, and test whether a larger NVLink domain justifies the operational requirements.

Updated Sep 2026 · Learn how AI infrastructure works through explanations, worked examples and diagrams, then practise applying it to interview questions.

Compare 64 or 72 useful GPUs, identify the communication that could improve, and test whether a larger NVLink domain justifies the operational requirements.

The full answer is part of Premium. A free account includes more answers in each topic, but does not unlock this one.

20 answers per topic instead of 10, plus saved progress and bookmarks · no cardor unlock all 316 remaining answers · ₹2,000 / $25

The concepts behind this question

Use these concept explanations to review the ideas behind this question.

🖧 Hardware & Cluster Build-Out

SXM, PCIe and Rack-Scale Form FactorsGPU form factors determine how accelerators connect, receive power and get serviced. Compare PCIe cards, SXM modules and rack-scale systems against the workload and the facility.

🖧 Hardware & Cluster Build-Out

NVLink Domains and the NVL72 RackAn NVLink domain connects a group of GPUs through a scale-up fabric. Its boundary helps determine where to place communication-heavy parallel groups.

🗂️ Scheduling & Orchestration🔒 Premium

Topology-Aware SchedulingTopology-aware scheduling places jobs with the communication network in mind. The right GPU count is not enough when ranks share slow or congested paths.

🧩 GPU & Accelerator Architecture🔒 Premium

NVLink, NVSwitch and PCIeNVLink, NVSwitch and PCIe provide different GPU communication paths. Learn how their topology and bandwidth affect the placement of parallel work.

Explore the full curriculum →

UP NEXT ON YOUR JOURNEY

Next in this trackBuy hardware, colocate, or rent from a GPU cloud? Work the decision for a 512-GPU need.Next in this trackYou have exactly 1 MW of facility capacity. Maximize training throughput.Next in this trackWhat does nvidia-fabricmanager do, and what exactly breaks when it is not running?

FEDITOR'S NOTE

Look for consistent bandwidth units, measured application benefit, equal-capacity comparisons and realistic failure boundaries.

DISCUSSION · 0

No comments yet — be the first to share your approach.