← Hardware, Cabling & Cluster Build-Out / 15

hardCrusoeCoreWeaveLambda Labs

Design the out-of-band management network for a 512-GPU cluster. What connects to it?

Design a management path that survives specified production failures. Count all endpoint classes, restrict device authority and test the complete recovery route.

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

Design a management path that survives specified production failures. Count all endpoint classes, restrict device authority and test the complete recovery route.

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.

🗂️ Scheduling & Orchestration🔒 Premium

Workload Identity and SecretsA running model service needs permission to read weights and emit telemetry. Separate the workload identity from its credentials, scope access to specific operations, and test expiry, revocation and cross-tenant denial.

🗂️ Scheduling & Orchestration🔒 Premium

GPU Tenant Isolation BoundariesGPU allocation, virtual machines and network policies protect different boundaries. Trace each tenant's device, network and storage paths, then test both permitted work and forbidden access before declaring a shared GPU service isolated.

🖧 Hardware & Cluster Build-Out

Burn-In and Acceptance TestingAcceptance testing checks that a new cluster meets agreed performance and reliability criteria. Sustained tests and comparisons between nodes help reveal defects before production.

Explore the full curriculum →

UP NEXT ON YOUR JOURNEY

Next in this trackAir, rear-door heat exchanger or direct-to-chip? Decide for a 60 kW rack.Next in this trackBuild the schedule for a 2,048-GPU deployment. What is on the critical path?Next in this trackHow many spare GPUs, nodes, cables and transceivers do you hold for a 2,048-GPU fleet?

FEDITOR'S NOTE

Look for physical port and failure mapping, independent service dependencies, controlled privileges and explicit test evidence.

DISCUSSION · 0

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