NCA - AI Infrastructure and Operations NCA-AIIO Practice Test 13
Question 120
An operations team implements: Consider a recommendation model trained on consented interaction data and evaluate ranking quality. Which test result would MOST convincingly show the approach works?
- The validation plan measures scenario coverage and compares simulator and real-world held-out performance.
- A test set verifies citations against the actual knowledge articles and escalation of unsupported requests.
- Evaluation on consented noisy calls reports word error rate and appropriate retention controls.
- An offline ranking benchmark and controlled experiment compare relevance and business outcomes.
Explanation
Correct Answer
D:The strongest proof is: An offline ranking benchmark and controlled experiment compare relevance and business outcomes. It directly tests the safeguard to consider a recommendation model trained on consented interaction data and evaluate ranking quality. The control addresses this specific context: A retailer wants to recommend products based on customers' prior interactions and item characteristics. It matters because recommendation systems predict relevance or preference rather than simply classifying one fixed label. Without it, poor personalization can reduce conversion while sensitive customer data are handled incorrectly.
Incorrect Answers
A:This observation would validate a different control: Use high-fidelity simulation or synthetic data generation with realism and generalization testing. It fits the case where an autonomous vehicle company trains a vision policy and needs rare-road-condition examples without public-road risk. It cannot prove success here; instead the required evidence is an offline ranking benchmark and controlled experiment compare relevance and business outcomes.
B:This observation would validate a different control: Combine retrieval-grounded generation with source attribution and a human escalation policy. It fits the case where a customer-service team wants an LLM to draft responses using approved knowledge-base articles. It cannot prove success here; instead the required evidence is an offline ranking benchmark and controlled experiment compare relevance and business outcomes.
C:This observation would validate a different control: Use automatic speech recognition with language coverage, privacy controls and transcription quality checks. It fits the case where a contact center wants automatic transcription of customer calls for searchable support records. It cannot prove success here; instead the required evidence is an offline ranking benchmark and controlled experiment compare relevance and business outcomes.
Question 121
The proposed approach is to add appropriate environmental sensors and integrate alerts with owned operational response procedures. In the following setting, a new AI room has reliable cooling capacity but no monitoring for localized humidity or leak alarms. Which principle BEST supports the approach?
- Physical maintainability and cooling access are mandatory facility design constraints.
- Facility monitoring should detect environmental conditions that threaten accelerated equipment operation.
- Facility availability depends on the complete power chain, not only server dual power supplies.
- Facility cabling must respect physical limits and support reliable future fault identification.
Explanation
Correct Answer
B:The governing principle is that facility monitoring should detect environmental conditions that threaten accelerated equipment operation. It supports the measure to add appropriate environmental sensors and integrate alerts with owned operational response procedures. because a new AI room has reliable cooling capacity but no monitoring for localized humidity or leak alarms. Without the control, a leak or humidity excursion may damage systems before a routine service inspection. The most useful acceptance evidence is: A controlled sensor test produces an actionable alarm with rack location and assigned response owner.
Incorrect Answers
A:That principle is relevant to this distinct case: A data center orders accelerated racks with rear-door heat exchangers but leaves insufficient service clearance. The associated response would be to check rack access, airflow and maintenance clearances before the final deployment layout. It does not describe the central reason for this case, which is that facility monitoring should detect environmental conditions that threaten accelerated equipment operation.
C:That principle is relevant to this distinct case: A GPU hall relies on one unprotected utility feed although the business requires continued operation through feed faults. The associated response would be to assess electrical redundancy, UPS autonomy and generator integration against the site's availability objective. It does not describe the central reason for this case, which is that facility monitoring should detect environmental conditions that threaten accelerated equipment operation.
D:That principle is relevant to this distinct case: A new cluster has redundant high-speed links but cable trays cannot support the planned fiber and copper bundles. The associated response would be to design documented cable routes, bend radii and labeling for power and network infrastructure. It does not describe the central reason for this case, which is that facility monitoring should detect environmental conditions that threaten accelerated equipment operation.
Question 122
The selected AI infrastructure safeguard is not implemented: Collect supported ECC error counters and evaluate hardware-health escalation thresholds. The observed situation is: A GPU node has increasing corrected memory errors but operations waits for application crashes before investigating. What exposure remains?
- An emerging device fault may remain undetected until a later uncorrectable error disrupts training.
- Localized GPU faults may persist for days while repetitive manual checks miss degraded nodes.
- A shared accelerator may meet aggregate throughput while violating individual service guarantees.
- Training can miss delivery targets while dashboards misinterpret throttled hardware as healthy.
Explanation
Correct Answer
A:The primary exposure is that an emerging device fault may remain undetected until a later uncorrectable error disrupts training. This follows from a GPU node has increasing corrected memory errors but operations waits for application crashes before investigating. where the omitted measure was to collect supported ECC error counters and evaluate hardware-health escalation thresholds. The technical reason is that corrected memory errors can provide early hardware-risk evidence even before visible job failure. A good operational validation after repair would be: An injected or simulated ECC warning reaches hardware maintenance with the affected GPU identifier.
Incorrect Answers
B:This exposure arises in a different situation: A fleet has thousands of accelerators but operators manually run nvidia-smi on each node to identify unhealthy GPUs. There, the corresponding safeguard is to use NVIDIA DCGM monitoring and supported telemetry exporters for centralized device diagnostics. In the present case, the more direct risk is that an emerging device fault may remain undetected until a later uncorrectable error disrupts training.
C:This exposure arises in a different situation: An inference service meets average latency but occasional requests become slow when multiple GPU tenants share the node. There, the corresponding safeguard is to correlate per-GPU compute, memory, scheduling contention and request tail latency. In the present case, the more direct risk is that an emerging device fault may remain undetected until a later uncorrectable error disrupts training.
D:This exposure arises in a different situation: An accelerator runs at reduced clock speed during a full training epoch but device utilization appears high. There, the corresponding safeguard is to correlate temperature, clock state and thermal-throttling reasons during sustained workload measurements. In the present case, the more direct risk is that an emerging device fault may remain undetected until a later uncorrectable error disrupts training.
Question 123
The team recommends this safeguard: Design a hybrid model lifecycle with approved artifact replication and separate sensitive-data access rules. It applies because a regulated enterprise trains models locally but wants cloud inference for geographically distributed customers. Which concept justifies the design?
- Workload criticality and recoverability should shape capacity purchasing choices.
- Regulatory and data-control constraints can override nominal per-GPU cloud price advantages.
- Infrastructure cost modeling must count data movement and persistence, not only accelerator rental time.
- Hybrid deployments can separate training and serving but still require trustworthy provenance and data governance.
Explanation
Correct Answer
D:The governing principle is that hybrid deployments can separate training and serving but still require trustworthy provenance and data governance. It supports the measure to design a hybrid model lifecycle with approved artifact replication and separate sensitive-data access rules. because a regulated enterprise trains models locally but wants cloud inference for geographically distributed customers. Without the control, an unreviewed model artifact or customer dataset may cross a security boundary during deployment. The most useful acceptance evidence is: A controlled promotion records model identity and proves inference receives only permitted data in each region.
Incorrect Answers
A:That principle is relevant to this distinct case: An AI service has predictable daily demand but a training task can be interrupted and resumed later. The associated response would be to select stable serving capacity and an appropriate lower-cost interruptible training tier with checkpoint recovery. It does not describe the central reason for this case, which is that hybrid deployments can separate training and serving but still require trustworthy provenance and data governance.
B:That principle is relevant to this distinct case: A government agency trains on restricted records that cannot leave an accredited local facility. The associated response would be to select an on-premises or explicitly approved sovereign environment consistent with data-residency rules. It does not describe the central reason for this case, which is that hybrid deployments can separate training and serving but still require trustworthy provenance and data governance.
C:That principle is relevant to this distinct case: A cloud-based training project moves large model checkpoints between regions and ignores transfer charges. The associated response would be to include data egress, cross-region transfer and retained storage costs in the deployment decision. It does not describe the central reason for this case, which is that hybrid deployments can separate training and serving but still require trustworthy provenance and data governance.
Question 124
The team recommends this safeguard: Select a fraud-scoring classifier with monitored precision, recall, drift and human escalation rules. It applies because a financial institution wants to flag unusual transaction patterns with tight limits on missed fraud and false alarms. Which concept justifies the design?
- Financial anomaly screening needs measured error trade-offs and controlled investigation decisions.
- Recommendation systems predict relevance or preference rather than simply classifying one fixed label.
- Grounded generative assistants should tie factual claims to approved content rather than rely only on model memory.
- Speech recognition converts audio to text and must account for noise, accents and domain-specific terms.
Explanation
Correct Answer
A:The governing principle is that financial anomaly screening needs measured error trade-offs and controlled investigation decisions. It supports the measure to select a fraud-scoring classifier with monitored precision, recall, drift and human escalation rules. because a financial institution wants to flag unusual transaction patterns with tight limits on missed fraud and false alarms. Without the control, a naive accuracy metric can hide rare fraudulent events within a largely legitimate dataset. The most useful acceptance evidence is: A held-out transaction set reports precision, recall and investigation burden at the selected threshold.
Incorrect Answers
B:That principle is relevant to this distinct case: A retailer wants to recommend products based on customers' prior interactions and item characteristics. The associated response would be to consider a recommendation model trained on consented interaction data and evaluate ranking quality. It does not describe the central reason for this case, which is that financial anomaly screening needs measured error trade-offs and controlled investigation decisions.
C:That principle is relevant to this distinct case: A customer-service team wants an LLM to draft responses using approved knowledge-base articles. The associated response would be to combine retrieval-grounded generation with source attribution and a human escalation policy. It does not describe the central reason for this case, which is that financial anomaly screening needs measured error trade-offs and controlled investigation decisions.
D:That principle is relevant to this distinct case: A contact center wants automatic transcription of customer calls for searchable support records. The associated response would be to use automatic speech recognition with language coverage, privacy controls and transcription quality checks. It does not describe the central reason for this case, which is that financial anomaly screening needs measured error trade-offs and controlled investigation decisions.
Question 125
The selected AI infrastructure safeguard is not implemented: Use automatic speech recognition with language coverage, privacy controls and transcription quality checks. The observed situation is: A contact center wants automatic transcription of customer calls for searchable support records. What exposure remains?
- Plausible unsupported answers may provide incorrect policy guidance to customers.
- Transcript errors can misrepresent customer instructions while sensitive recordings remain exposed.
- Poor personalization can reduce conversion while sensitive customer data are handled incorrectly.
- An unvalidated prediction could be mistaken for a diagnosis and affect patient care.
Explanation
Correct Answer
B:The primary exposure is that transcript errors can misrepresent customer instructions while sensitive recordings remain exposed. This follows from a contact center wants automatic transcription of customer calls for searchable support records. where the omitted measure was to use automatic speech recognition with language coverage, privacy controls and transcription quality checks. The technical reason is that speech recognition converts audio to text and must account for noise, accents and domain-specific terms. A good operational validation after repair would be: Evaluation on consented noisy calls reports word error rate and appropriate retention controls.
Incorrect Answers
A:This exposure arises in a different situation: A customer-service team wants an LLM to draft responses using approved knowledge-base articles. There, the corresponding safeguard is to combine retrieval-grounded generation with source attribution and a human escalation policy. In the present case, the more direct risk is that transcript errors can misrepresent customer instructions while sensitive recordings remain exposed.
C:This exposure arises in a different situation: A retailer wants to recommend products based on customers' prior interactions and item characteristics. There, the corresponding safeguard is to consider a recommendation model trained on consented interaction data and evaluate ranking quality. In the present case, the more direct risk is that transcript errors can misrepresent customer instructions while sensitive recordings remain exposed.
D:This exposure arises in a different situation: A radiology team needs to highlight suspicious structures in diagnostic images for qualified human review. There, the corresponding safeguard is to evaluate computer-vision segmentation with clinically governed validation and human oversight. In the present case, the more direct risk is that transcript errors can misrepresent customer instructions while sensitive recordings remain exposed.
Question 126
Consider this NVIDIA AI infrastructure scenario: A high-performance computing group trains large models requiring fast node-to-node collective operations. Which design choice should be made FIRST?
- Select supported optics, fiber media and loss budgets for the required data-center link distances.
- Assess a tuned high-speed Ethernet or RoCE solution rather than assuming ordinary LAN defaults are enough.
- Evaluate a suitably sized InfiniBand fabric with documented latency, bandwidth and management requirements.
- Use supported NVLink or NVSwitch paths for local GPU exchange rather than assuming every copy crosses Ethernet.
Explanation
Correct Answer
C:The appropriate response is to evaluate a suitably sized InfiniBand fabric with documented latency, bandwidth and management requirements. This is specific to the case where a high-performance computing group trains large models requiring fast node-to-node collective operations. The mechanism is that high-performance InfiniBand networks can support latency-sensitive collective traffic between AI nodes. If ignored, a fabric not designed for collectives may make distributed scaling inefficient. Validate the choice using this targeted evidence: Measured collective performance meets the training job's communication budget across the intended node count.
Incorrect Answers
A:This alternative addresses a different requirement: A planned GPU cluster spans distant data halls but its optical cabling design ignores reach and transceiver compatibility. It would be appropriate there to select supported optics, fiber media and loss budgets for the required data-center link distances. But the present scenario is about a high-performance computing group trains large models requiring fast node-to-node collective operations. For it, the needed control is to evaluate a suitably sized InfiniBand fabric with documented latency, bandwidth and management requirements.
B:This alternative addresses a different requirement: A cloud provider standardizes on Ethernet switching but wants accelerated AI traffic over the existing operational model. It would be appropriate there to assess a tuned high-speed Ethernet or RoCE solution rather than assuming ordinary LAN defaults are enough. But the present scenario is about a high-performance computing group trains large models requiring fast node-to-node collective operations. For it, the needed control is to evaluate a suitably sized InfiniBand fabric with documented latency, bandwidth and management requirements.
D:This alternative addresses a different requirement: A scale-up system exchanges large tensors between GPUs inside one chassis and compares that traffic with rack networking. It would be appropriate there to use supported NVLink or NVSwitch paths for local GPU exchange rather than assuming every copy crosses Ethernet. But the present scenario is about a high-performance computing group trains large models requiring fast node-to-node collective operations. For it, the needed control is to evaluate a suitably sized InfiniBand fabric with documented latency, bandwidth and management requirements.
Question 127
The proposed approach is to evaluate supported GPUDirect RDMA and the required GPU, NIC, driver and topology compatibility. In the following setting, a research team expects GPUs to access distributed data without unnecessary staging through host CPU memory. Which principle BEST supports the approach?
- Direct accelerator data paths can reduce host-memory copies when the complete platform supports them.
- NIC selection can limit effective GPU cluster throughput despite ample internal device bandwidth.
- Network isolation limits unauthorized access and protects operational control from workload contention.
- Distributed AI networks must be sized for realistic many-to-many communication patterns.
Explanation
Correct Answer
A:The governing principle is that direct accelerator data paths can reduce host-memory copies when the complete platform supports them. It supports the measure to evaluate supported GPUDirect RDMA and the required GPU, NIC, driver and topology compatibility. because a research team expects GPUs to access distributed data without unnecessary staging through host CPU memory. Without the control, unsupported assumptions may cause fallback transfers and unexpected CPU overhead. The most useful acceptance evidence is: A validated supported path records lower host-copy overhead with correct end-to-end data delivery.
Incorrect Answers
B:That principle is relevant to this distinct case: An accelerated node uses a slow network interface although its GPU data path needs high-bandwidth peer transfers. The associated response would be to select supported high-speed NIC or HCA capacity aligned with expected accelerator and storage traffic. It does not describe the central reason for this case, which is that direct accelerator data paths can reduce host-memory copies when the complete platform supports them.
C:That principle is relevant to this distinct case: A shared GPU cluster combines tenant inference traffic and privileged management access on one unmanaged network. The associated response would be to segment workload and management paths with required access control and monitored boundaries. It does not describe the central reason for this case, which is that direct accelerator data paths can reduce host-memory copies when the complete platform supports them.
D:That principle is relevant to this distinct case: A leaf-spine AI network uses heavily oversubscribed uplinks and passes only simple single-node benchmarks. The associated response would be to calculate fabric bisection bandwidth and oversubscription for the planned scale-out job mix. It does not describe the central reason for this case, which is that direct accelerator data paths can reduce host-memory copies when the complete platform supports them.
Question 128
In this situation, a newly promoted inference model produces biased results and no earlier artifact or configuration was retained. The team does not maintain a tested rollback path with previous model artifacts, endpoints and compatibility metadata. Which consequence BEST characterizes the gap?
- The team may be unable to restore reliable predictions quickly after a problematic release.
- An accurate checkpoint can become unavailable or break client requests after release.
- A faulty input batch can cause an inferior model to replace a stable production version.
- Silent drift can degrade decisions while system CPU and GPU health still look normal.
Explanation
Correct Answer
A:The primary exposure is that the team may be unable to restore reliable predictions quickly after a problematic release. This follows from a newly promoted inference model produces biased results and no earlier artifact or configuration was retained. where the omitted measure was to maintain a tested rollback path with previous model artifacts, endpoints and compatibility metadata. The technical reason is that model recovery requires preserving approved binaries and the serving configuration needed to restore them. A good operational validation after repair would be: An induced regression triggers a rollback to a known prior model and verifies representative outputs.
Incorrect Answers
B:This exposure arises in a different situation: A model-serving change reaches production without checking its API schema, resource limits or startup health. There, the corresponding safeguard is to use staged deployment validation for model artifact, endpoint compatibility, GPU resources and readiness. In the present case, the more direct risk is that the team may be unable to restore reliable predictions quickly after a problematic release.
C:This exposure arises in a different situation: An organization automatically retrains a model nightly and publishes weights without independent acceptance rules. There, the corresponding safeguard is to gate retraining output on dataset validation, safety metrics and approved promotion criteria. In the present case, the more direct risk is that the team may be unable to restore reliable predictions quickly after a problematic release.
D:This exposure arises in a different situation: A demand-forecasting model loses accuracy months after deployment as user behavior changes. There, the corresponding safeguard is to monitor production data distribution, predictive performance and drift with assigned retraining triggers. In the present case, the more direct risk is that the team may be unable to restore reliable predictions quickly after a problematic release.
Question 129
The proposed approach is to inspect supported NVLink status and error counters alongside application collective performance. In the following setting, multi-GPU jobs slow after cabling work and a topology view shows one NVLink path with elevated errors. Which principle BEST supports the approach?
- GPU utilization is one diagnostic indicator and does not uniquely identify the performance bottleneck.
- Power-cap telemetry helps distinguish policy-limited compute throughput from ordinary algorithm inefficiency.
- Xid diagnostics provide context for GPU software or hardware failures and require supported interpretation.
- GPU interconnect health directly influences distributed work within supported accelerator systems.
Explanation
Correct Answer
D:The governing principle is that gPU interconnect health directly influences distributed work within supported accelerator systems. It supports the measure to inspect supported NVLink status and error counters alongside application collective performance. because multi-GPU jobs slow after cabling work and a topology view shows one NVLink path with elevated errors. Without the control, a degraded link can add latency while individual GPU compute metrics look healthy. The most useful acceptance evidence is: A link-health check identifies the affected path and an approved repair restores collective time.
Incorrect Answers
A:That principle is relevant to this distinct case: A GPU has low reported utilization while its training job is slow and the team assumes the accelerator is defective. The associated response would be to correlate SM utilization with CPU data loading, communication time and memory bandwidth measures. It does not describe the central reason for this case, which is that gPU interconnect health directly influences distributed work within supported accelerator systems.
B:That principle is relevant to this distinct case: A data center plans more training workers despite existing GPUs regularly hitting enforced power limits. The associated response would be to monitor board power draw, configured limits and resulting effective clock rates before expanding hardware. It does not describe the central reason for this case, which is that gPU interconnect health directly influences distributed work within supported accelerator systems.
C:That principle is relevant to this distinct case: A containerized training job unexpectedly exits and system logs contain GPU Xid error messages. The associated response would be to correlate NVIDIA Xid events with driver status, hardware health and the affected job timeline. It does not describe the central reason for this case, which is that gPU interconnect health directly influences distributed work within supported accelerator systems.
Question 130
Consider this NVIDIA AI infrastructure scenario: A multi-GPU training model requires full-GPU memory but a manager proposes partitioning every GPU into tiny MIG slices. Which design choice should be made FIRST?
- Distinguish node-local NVLink or NVSwitch scaling from networked multi-node scale-out design.
- Reserve suitable full GPU devices for workloads that need whole-device capacity or communication topology.
- Model aggregate east-west traffic and network oversubscription at the intended cluster scale.
- Use an appropriate workload scheduler with GPU resource accounting and topology-aware placement.
Explanation
Correct Answer
B:The appropriate response is to reserve suitable full GPU devices for workloads that need whole-device capacity or communication topology. This is specific to the case where a multi-GPU training model requires full-GPU memory but a manager proposes partitioning every GPU into tiny MIG slices. The mechanism is that gPU partitioning is a workload-isolation choice, not an automatic substitute for full-device training capacity. If ignored, an otherwise valid model may be unable to fit in a slice or access the intended multi-GPU topology. Validate the choice using this targeted evidence: Scheduling tests place full-device training jobs and separate eligible small inference slices correctly.
Incorrect Answers
A:This alternative addresses a different requirement: A team adds more GPUs inside one system and assumes the same link topology extends to GPUs in other racks. It would be appropriate there to distinguish node-local NVLink or NVSwitch scaling from networked multi-node scale-out design. But the present scenario is about a multi-GPU training model requires full-GPU memory but a manager proposes partitioning every GPU into tiny MIG slices. For it, the needed control is to reserve suitable full GPU devices for workloads that need whole-device capacity or communication topology.
C:This alternative addresses a different requirement: A data center scales from four to sixty-four training nodes without upgrading per-node network ports. It would be appropriate there to model aggregate east-west traffic and network oversubscription at the intended cluster scale. But the present scenario is about a multi-GPU training model requires full-GPU memory but a manager proposes partitioning every GPU into tiny MIG slices. For it, the needed control is to reserve suitable full GPU devices for workloads that need whole-device capacity or communication topology.
D:This alternative addresses a different requirement: A shared accelerator fleet has idle GPUs in some nodes while large training jobs wait without proper scheduling. It would be appropriate there to use an appropriate workload scheduler with GPU resource accounting and topology-aware placement. But the present scenario is about a multi-GPU training model requires full-GPU memory but a manager proposes partitioning every GPU into tiny MIG slices. For it, the needed control is to reserve suitable full GPU devices for workloads that need whole-device capacity or communication topology.