NVIDIA NCP-AII Questions & Answers

The related but distinct situation is “A BCM operator thinks the named primary head must always be active, even after the secondary has taken over successfully”. The appropriate intervention is Distinguish fixed primary/secondary identity from changing active/passive operational roles in HA validation. For the stated problem, “Install and verify supported primary and secondary BCM head-node high availability with safe ownership transfer” addresses the central condition.

This option serves the scenario “Two BCM head nodes both attempt to mount a shared writable filesystem during a simulated split-brain incident”. The direct fix is Configure supported HA fencing, quorum and shared-resource ownership protections before approving automated failover. The present scenario is distinguished by “Install and verify supported primary and secondary BCM head-node high availability with safe ownership transfer”.

This alternative addresses the separate case “A network partition causes a passive BCM head to see the active head as unreachable, and an operator immediately promotes it without quorum checks”. The appropriate intervention is Use the documented BCM HA quorum and split-brain prevention procedure before changing service ownership. The deciding requirement here is “Install and verify supported primary and secondary BCM head-node high availability with safe ownership transfer”.

This is weaker evidence because Visible NVIDIA devices do not guarantee that the application framework contains GPU-enabled binaries. The independent test must show that The framework reports the expected accelerator device and a measured operation uses GPU utilization.

This option serves the scenario “A Docker workload requires only one accelerator, but the launch command passes all eight GPUs from its shared host”. The decisive evidence is that A container inventory displays only the single allocated GPU UUID and the workload passes its test. This question instead requires “The framework reports the expected accelerator device and a measured operation uses GPU utilization”.

The candidate evidence checks Run a user- and image-specific GPU smoke test with the actual deployment security context in another context. Here the specific acceptance result is The framework reports the expected accelerator device and a measured operation uses GPU utilization.

This alternative addresses the separate case “An engineer installs an optic that matches the physical cage but uses a wavelength incompatible with the remote transceiver”. The operational evidence follows from the mechanism that Matching connector geometry does not imply optical signaling compatibility between transceiver endpoints. The deciding requirement here is “Corrected FEC events can reveal deteriorating signal margin before hard link failures occur”.

The option would apply to “A 400G fabric port uses a short passive copper cable beyond its approved reach and link quality is unstable”. The operational evidence follows from the mechanism that Passive copper, active copper and optical transceivers have distinct reach and signal-integrity envelopes. This question instead requires “Corrected FEC events can reveal deteriorating signal margin before hard link failures occur”.

The related but distinct situation is “A dual-rail rack has all high-speed ports connected but two cables are accidentally crossed between fabrics”. The operational evidence follows from the mechanism that Correct link state does not guarantee that each cable terminates on the correct rail or security domain. For the stated problem, “Corrected FEC events can reveal deteriorating signal margin before hard link failures occur” addresses the central condition.

This alternative addresses the separate case “During HPL burn-in one GPU's temperature and power-limit counters increase but the aggregate benchmark still finishes”. The primary exposure is that Leaving a hot accelerator in service can cause production throttling or failure during longer jobs. For the stated problem, “Ignoring intermittent arithmetic validation failures can expose training to silent data corruption” addresses the central condition.

The related but distinct situation is “A one-minute HPL run is used for a rack acceptance contract that requires proving sustained computation under design power”. The primary exposure is that Prolonged compute stress may reveal faults that were absent during the initial quick score. The deciding requirement here is “Ignoring intermittent arithmetic validation failures can expose training to silent data corruption”.

This option serves the scenario “An engineer compares two long HPL burn-in periods that use different matrix dimensions and CUDA libraries without documenting the difference”. The acceptance board should flag that Version and workload drift can be misread as a hardware degradation or recovery. This question instead requires “Ignoring intermittent arithmetic validation failures can expose training to silent data corruption”.

Although the alternative can be true, the current evidence is more directly governed by An optimal storage layout depends on workload I/O size and concurrency, not a universal maximum stripe count.

The related but distinct situation is “A shared storage server reaches a high benchmark throughput but training hosts still see long I/O stalls because their mount and client NIC settings differ”. The key distinction is that A storage-system benchmark alone does not establish delivered I/O performance at the application host. For the stated problem, “An optimal storage layout depends on workload I/O size and concurrency, not a universal maximum stripe count” addresses the central condition.

This alternative addresses the separate case “A training job streams datasets from a shared NFS path and repeatedly rereads the same small files over an overloaded link”. The operational evidence follows from the mechanism that Data-loading performance depends on access pattern and locality as well as storage peak bandwidth. The deciding requirement here is “An optimal storage layout depends on workload I/O size and concurrency, not a universal maximum stripe count”.

This option addresses an alternative scenario (ClusterKit reports a slow InfiniBand fabric but operators run it during a large unrelated backup and treat the result as hardware failure); the reported issue instead calls for Run ClusterKit's appropriate multifaceted cluster assessment in addition to individual benchmark tests.

The suggested action targets another problem where Aggregate scores can hide a fundamental failed infrastructure dependency among otherwise healthy tests. It does not directly address this case, which needs Run ClusterKit's appropriate multifaceted cluster assessment in addition to individual benchmark tests.

The option would apply to “A new cluster's ClusterKit run omits three unreachable nodes from the result table and the engineer marks the entire fleet as passed”. The direct fix is Reconcile benchmark participants against the complete intended cluster node inventory. The present scenario is distinguished by “Run ClusterKit's appropriate multifaceted cluster assessment in addition to individual benchmark tests”.

This option serves the scenario “A cluster node has an appropriate GPU driver but a signed kernel refuses to load the unsigned NVIDIA module under secure boot”. The appropriate intervention is Install and sign the required driver module using approved key enrollment procedures. The present scenario is distinguished by “Drain the node and perform the supported module activation procedure before acceptance”.

This option addresses an alternative scenario (A technician uninstalls the NVIDIA driver without checking which running jobs and container images depend on it); the reported issue instead calls for Drain the node and perform the supported module activation procedure before acceptance.

The related but distinct situation is “A host updates the CUDA toolkit but leaves an incompatible NVIDIA kernel driver and the application fails to initialize CUDA”. Therefore Align the installed GPU driver and userspace software with NVIDIA's supported compatibility information. For the stated problem, “Drain the node and perform the supported module activation procedure before acceptance” addresses the central condition.

This alternative addresses the separate case “A NeMo burn-in has good GPU utilization but model checkpoints intermittently fail because shared storage fills during long runs”. The acceptance board should flag that Unobserved checkpoint storage exhaustion can discard hours of otherwise healthy training. For the stated problem, “Synthetic-only smoke tests can miss application-stack integration failures at production duration” addresses the central condition.

The option would apply to “A NeMo training burn-in resumes from a checkpoint but silently changes tokenizer and model configuration between runs”. Without correction, An inconsistent resume can invalidate benchmark comparisons and corrupt model quality. The present scenario is distinguished by “Synthetic-only smoke tests can miss application-stack integration failures at production duration”.

This option serves the scenario “A NeMo benchmark appears slow while GPU and network metrics remain normal but tokenization input workers are pegged at CPU limits”. The acceptance board should flag that Misdiagnosing CPU preprocessing as GPU failure can lead to unnecessary accelerator replacement. This question instead requires “Synthetic-only smoke tests can miss application-stack integration failures at production duration”.

The option would apply to “A failed network adapter is replaced but the cluster job scheduler still references the prior interface and HCA identity”. The key distinction is that New hardware may enumerate differently and invalidate previously pinned interface or RDMA selections. This question instead requires “Replacing accelerator hardware is a controlled maintenance operation, not a live in-band software fix”.

The related but distinct situation is “A replacement PSU has the same mechanical shape but a lower supported power rating than the original module”. The operational evidence follows from the mechanism that Mechanical fit does not imply the module supports the electrical envelope of a high-power GPU chassis. For the stated problem, “Replacing accelerator hardware is a controlled maintenance operation, not a live in-band software fix” addresses the central condition.

This alternative addresses the separate case “A technician replaces a GPU successfully but does not reapply the workload partition profile required for the tenant scheduler”. This case illustrates that A hardware swap can reset logical GPU resources and alter device identities used by jobs. The deciding requirement here is “Replacing accelerator hardware is a controlled maintenance operation, not a live in-band software fix”.

The related but distinct situation is “A BlueField-3 firmware update reports success but its active image is still the old slot after restart”. The acceptance board should flag that Treating a staged image as active can leave the fleet with inconsistent capabilities. The deciding requirement here is “Mistaking intentional host restrictions for defects can trigger unnecessary DPU replacement”.

This option serves the scenario “BlueField-3 adapters pass host Ethernet traffic but the embedded Arm software revision differs from the approved fabric interoperability matrix”. Without correction, An unnoticed embedded firmware mismatch can trigger offload faults under distributed workloads. This question instead requires “Mistaking intentional host restrictions for defects can trigger unnecessary DPU replacement”.

This alternative overstates a separate issue; the current case should be prioritized for the risk that Mistaking intentional host restrictions for defects can trigger unnecessary DPU replacement.

This option serves the scenario “Two otherwise identical optical links use transceiver firmware revisions that negotiate different advertised capabilities on the same switch”. The acceptance board should flag that Unreviewed optic firmware differences can cause subtle link training regressions. The present scenario is distinguished by “Ignoring optics in the firmware baseline can keep known intermittent link problems unresolved”.

The option would apply to “A replacement transceiver has the correct vendor SKU but an old firmware image that fails supported link diagnostics”. The acceptance board should flag that Outdated optic firmware can reduce diagnostic coverage or prevent reliable link setup. This question instead requires “Ignoring optics in the firmware baseline can keep known intermittent link problems unresolved”.

The related but distinct situation is “An engineer replaces a failed optical module and assumes no post-replacement tests are necessary because the link LED is green”. The acceptance board should flag that A marginal replacement can fail later under higher rack temperature or sustained bandwidth. For the stated problem, “Ignoring optics in the firmware baseline can keep known intermittent link problems unresolved” addresses the central condition.