← Hardware, Cabling & Cluster Build-Out / 12

mediumCrusoeCoreWeaveNVIDIA

Count the switches, cables and transceivers for a 2,048-GPU rail-optimized fabric.

Count one complete rail, then distinguish logical links from cages, optical modules and cable assemblies. Derive power from the same physical inventory.

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

Count one complete rail, then distinguish logical links from cages, optical modules and cable assemblies. Derive power from the same physical inventory.

This answer is included with a free account. Sign in to read it and save your progress.

free account · no card needed

The concepts behind this question

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

🖧 Hardware & Cluster Build-Out

The Bill of Materials for a Training ClusterA cluster bill of materials includes networking, storage, management, power, cooling and spares as well as GPUs. Size each from the workload and topology.

🖧 Hardware & Cluster Build-Out

Scale-Out Fabric Choice: InfiniBand XDR vs Spectrum-XInfiniBand and Spectrum-X connect GPUs across nodes. Compare congestion behavior, topology and operational requirements as well as port speed.

🔌 Networking & Storage🔒 Premium

Rail-Optimized and Fat-Tree FabricsRail-optimized and Clos fabrics organize traffic between GPU nodes. Port counts, placement and oversubscription determine which paths are available to collectives.

🖧 Hardware & Cluster Build-Out

Cables, Transceivers and the Optics Power BudgetCable selection depends on distance, port compatibility and power. Count both ends of optical links and check the exact part numbers before ordering.

Explore the full curriculum →

UP NEXT ON YOUR JOURNEY

Next in this trackAfter a maintenance window, two-node bandwidth halved. Single-node is unchanged. Diagnose it.Next in this trackSize the storage for a 2,048-GPU training cluster. What numbers actually drive it?Next in this trackDesign the out-of-band management network for a 512-GPU cluster. What connects to it?

FEDITOR'S NOTE

Look for feasible leaf/spine wiring, port conservation, rail-specific growth and correct treatment of twin-port optics.

DISCUSSION · 0

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