NVIDIA NCP-AII Questions & Answers

The option would apply to “ClusterKit reports a slow InfiniBand fabric but operators run it during a large unrelated backup and treat the result as hardware failure”. The primary exposure is that Uncontrolled test timing can trigger needless component replacement and unreliable acceptance decisions. This question instead requires “Narrow acceptance can miss fabric or node configuration defects before customer workloads begin”.

This option serves the scenario “A ClusterKit job fails to execute on a subset of nodes due to missing execution permissions and is reported as poor GPU performance”. The direct failure mode is that Treating missing benchmark prerequisites as performance evidence can misdirect troubleshooting. The present scenario is distinguished by “Narrow acceptance can miss fabric or node configuration defects before customer workloads begin”.

This alternative addresses the separate case “An operator averages all ClusterKit results into one green score even though one test suite failed every storage path in a rack”. The direct failure mode is that A broad average can grant acceptance to a cluster that cannot read its training dataset. The deciding requirement here is “Narrow acceptance can miss fabric or node configuration defects before customer workloads begin”.

This option serves the scenario “A Base Command Manager cluster relies on one head node and loses both DHCP provisioning and Slurm scheduling whenever that node powers off”. Without correction, An untested single head node can become a cluster-wide control-plane outage. This question instead requires “A shared network or power dependency can neutralize the planned control-plane redundancy”.

This alternative addresses the separate case “Two BCM head nodes both attempt to mount a shared writable filesystem during a simulated split-brain incident”. The acceptance board should flag that Split brain can corrupt management and job metadata during apparent redundancy. For the stated problem, “A shared network or power dependency can neutralize the planned control-plane redundancy” addresses the central condition.

The option would apply to “A backup head node has no configured remote power-fencing capability, but the team expects automatic recovery from any loss of heartbeat”. The primary exposure is that Assuming unfenced automatic failover can leave a cluster unavailable or vulnerable to dual writers. The present scenario is distinguished by “A shared network or power dependency can neutralize the planned control-plane redundancy”.

This option serves the scenario “A technician uses an optical transceiver with unsupported insertion loss and the switch intermittently changes link speed”. The direct failure mode is that A mismatched optical channel can undermine network reliability even when the cable looks intact. The present scenario is distinguished by “Marginal signal quality can cause intermittent retraining and distributed collective timeouts”.

This alternative addresses the separate case “A new breakout cable passes continuity checks but its highest-speed lanes show repeated correctable errors under traffic”. The direct failure mode is that Marginal breakout lanes can create sporadic packet loss that is hard to reproduce in applications. The deciding requirement here is “Marginal signal quality can cause intermittent retraining and distributed collective timeouts”.

The related but distinct situation is “A high-speed NIC link shows error spikes only after adjacent power cables are rerouted tightly against the data assembly”. Without correction, Ignoring physical-layer handling can allow recurring fabric faults after routine facility changes. For the stated problem, “Marginal signal quality can cause intermittent retraining and distributed collective timeouts” addresses the central condition.

This alternative addresses the separate case “An operator runs an AI application on a fresh server but does not record the GPU driver version or actual compute device selection”. The direct failure mode is that Missing device and software mapping can conceal workload placement or compatibility defects. For the stated problem, “An undetected missing accelerator reduces planned capacity and can break distributed rank mapping” addresses the central condition.

The related but distinct situation is “A server shows NVIDIA GPUs in the PCI bus but nvidia-smi fails with a driver communication error”. The acceptance board should flag that Conflating PCI presence with driver readiness can lead to unusable accelerator nodes entering the scheduler. The deciding requirement here is “An undetected missing accelerator reduces planned capacity and can break distributed rank mapping”.

The alternative may occur in another situation (An HGX server has eight SXM accelerators but a runbook treats it as eight independent PCIe cards for service procedures), but the principal consequence here is that An undetected missing accelerator reduces planned capacity and can break distributed rank mapping.

The option would apply to “A collective burn-in reports consistent success but one rack contributes very low bus bandwidth and extends global completion time”. The direct failure mode is that An unaddressed straggler can degrade every training step even without explicit job failure. This question instead requires “Accepting one short collective run can miss recurring job-killing fabric instability”.

The related but distinct situation is “An NCCL burn-in is interrupted by a scheduled switch update and the report falsely labels the hardware rack as failing”. The acceptance board should flag that Uncoordinated burn-in and maintenance can create misleading failure investigations. For the stated problem, “Accepting one short collective run can miss recurring job-killing fabric instability” addresses the central condition.

This alternative addresses the separate case “A long NCCL test passes only with two ranks while production jobs regularly use 64 GPUs distributed across multiple racks”. The primary exposure is that A two-rank pass cannot certify distributed collectives at the scale used by customers. The deciding requirement here is “Accepting one short collective run can miss recurring job-killing fabric instability”.

This option serves the scenario “A dual-port NIC exposes only one working port after an optic swap and the operator treats the whole adapter as failed”. The appropriate intervention is Cross-check port-level firmware, transceiver and cable diagnostics before labeling the NIC defective. This question instead requires “Check PCI enumeration, slot power, cable seating and platform event logs for the affected adapter”.

This alternative addresses the separate case “A GPU is absent from the expected device inventory after maintenance, but the technician proposes reinstalling CUDA before checking hardware state”. Therefore Compare BMC inventory, lspci, physical seating and power connections to identify an actually missing device. For the stated problem, “Check PCI enumeration, slot power, cable seating and platform event logs for the affected adapter” addresses the central condition.

The option would apply to “A rack reports repeated fan alerts from one slot, but operators replace another fan with the same visible color label”. The direct fix is Map BMC slot identifiers and physical fan positions from the qualified chassis service documentation. The present scenario is distinguished by “Check PCI enumeration, slot power, cable seating and platform event logs for the affected adapter”.

This alternative addresses the separate case “Two tenants see the same MIG device assignment because an orchestration template reuses one accelerator identifier”. The primary exposure is that Duplicated MIG mapping can cause contention, failed starts or isolation violations. The deciding requirement here is “Ignoring validated partitioning can waste costly GPU capacity and weaken workload placement controls”.

The option would apply to “After reboot, a previously assigned MIG layout disappears and the scheduler tries to place jobs on nonexistent device UUIDs”. Without correction, An ephemeral MIG configuration can strand or misplace jobs after routine maintenance. This question instead requires “Ignoring validated partitioning can waste costly GPU capacity and weaken workload placement controls”.

The related but distinct situation is “An administrator enables MIG mode on a supported GPU but creates no GPU or compute instances and the workload sees no device”. The acceptance board should flag that A half-configured MIG host can be discovered but offer no usable accelerator capacity. For the stated problem, “Ignoring validated partitioning can waste costly GPU capacity and weaken workload placement controls” addresses the central condition.

This alternative addresses the separate case “An administrator installs the NVIDIA Container Toolkit but Docker workloads still show no GPUs because the runtime is not configured”. The strongest acceptance result is that A supported GPU-enabled container lists the expected devices after runtime configuration, not merely that a command executed. For the stated problem, “A standard GPU-container smoke test runs successfully after the daemon upgrade” addresses the central condition.

This is weaker evidence because Runtime integration is version-sensitive and can change independently of the installed host GPU driver. The independent test must show that A standard GPU-container smoke test runs successfully after the daemon upgrade.

The related but distinct situation is “A security baseline blocks device pass-through entirely even though authorized containers require selected GPU access”. A completed remediation must demonstrate that An authorized container sees its allowed GPU UUID while an unauthorized container sees none. The deciding requirement here is “A standard GPU-container smoke test runs successfully after the daemon upgrade”.

The option would apply to “An operator checks spine switches only by ping and does not inspect temperature, transceiver status or port error counters”. A completed remediation must demonstrate that The signed switch report records healthy sensors, expected interfaces and low error rates during burn-in. This question instead requires “Post-restart readback matches policy and NCCL fabric performance remains within threshold”.

The proposed validation does not isolate the reported fault. An effective closure test requires evidence that Post-restart readback matches policy and NCCL fabric performance remains within threshold.

This alternative addresses the separate case “A switch is reachable over SSH but the installed firmware does not support the configured high-speed port breakout profile”. The decisive evidence is that Switch inventory and active interface profiles match supported capabilities and all intended ports are operational. The deciding requirement here is “Post-restart readback matches policy and NCCL fabric performance remains within threshold”.

That change may help in a different context, but it would not establish that A sustained repeat workload completes without new GPU Xid or uncorrectable ECC events under supported limits. The relevant response is Run a controlled single-node accelerator stress and correlate Xid messages, ECC, thermals and power readings.

This option serves the scenario “A stress script records exit code zero even when a GPU has reported uncorrectable memory errors during the test”. The appropriate intervention is Correlate application status with management telemetry, kernel logs and accelerator hardware fault counters. This question instead requires “Run a controlled single-node accelerator stress and correlate Xid messages, ECC, thermals and power readings”.

This alternative addresses the separate case “An acceptance engineer benchmarks only GPU number zero on an eight-GPU HGX server and records the entire node as healthy”. Therefore Stress every installed accelerator simultaneously and reconcile per-GPU health and performance results. For the stated problem, “Run a controlled single-node accelerator stress and correlate Xid messages, ECC, thermals and power readings” addresses the central condition.

This option serves the scenario “An eight-GPU server is assigned to a PDU feed sized for average power rather than accelerator transient peaks”. The key distinction is that Accelerator power spikes and redundancy requirements must fit within the electrical distribution design. The present scenario is distinguished by “Air recirculation and bypass can create hot device inlets even when total chiller capacity is adequate”.

This alternative addresses the separate case “A GPU server draws 10 kW at full training load but the facilities acceptance test used only idle operating-system power”. This case illustrates that Idle measurements systematically understate AI compute thermal and electrical requirements. The deciding requirement here is “Air recirculation and bypass can create hot device inlets even when total chiller capacity is adequate”.

The option would apply to “A liquid-cooled AI server is powered before facility coolant flow and leak sensing are commissioned”. The key distinction is that Liquid cooling is an operational dependency of supported high-power accelerator hardware, not optional decoration. This question instead requires “Air recirculation and bypass can create hot device inlets even when total chiller capacity is adequate”.