NCP-AIO Workload Management Practice Question
An administrator is configuring a Kubernetes cluster where some nodes have A100 GPUs and others have H100 GPUs. A training job requires specific GPU memory capacity and CUDA compute capability. Which mechanism should the administrator use to ensure the job is only scheduled onto nodes with the correct GPU model?
⚠ Common exam trap
The trap here is treating nvidia.com/gpu as a model or memory selector, when it is only a device count and cannot distinguish between A100 and H100 hardware.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Set a nodeSelector matching the GPU model label applied by the GPU Operator's node feature discovery
Node Feature Discovery, part of the GPU Operator stack, labels nodes with GPU product details such as nvidia.com/gpu.product. A nodeSelector or affinity rule referencing that label restricts scheduling to nodes with the required GPU model, which is the precise way to satisfy memory and compute capability requirements without manual node naming.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Apply a taint to nodes without the required GPU and rely on the default scheduler
Why it's wrong here
Taints repel pods unless they tolerate them, but without a corresponding toleration on the training pod, the pod would avoid all tainted nodes, including valid ones. Taints alone do not select for a specific GPU model; they only exclude nodes. This approach is imprecise and does not reliably target the correct hardware.
- ✗
Increase the nvidia.com/gpu resource request to match the GPU memory size
Why it's wrong here
The nvidia.com/gpu resource represents a count of devices, not a memory size. Requesting more GPUs does not filter by model or memory capacity and may simply allocate multiple smaller GPUs. It cannot distinguish an A100 from an H100, so it fails to enforce the model-specific scheduling requirement.
- ✓
Set a nodeSelector matching the GPU model label applied by the GPU Operator's node feature discovery
Why this is correct
Node Feature Discovery, deployed with the GPU Operator, labels nodes with GPU model information such as nvidia.com/gpu.product. A nodeSelector referencing that label constrains the scheduler to nodes with the required GPU model, ensuring the job lands only on A100 or H100 nodes as appropriate without hardcoding node names.
- ✗
Use a ResourceQuota that limits GPU requests per namespace
Why it's wrong here
A ResourceQuota caps aggregate resource consumption within a namespace but does not influence which node a pod is placed on. It cannot express a preference for GPU model or compute capability, so it does nothing to prevent the training job from being scheduled onto an incompatible GPU node.
About these practice questions
Courseiva writes every NCP-AIO question from scratch — 309 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official NVIDIA exam blueprint
This NCP-AIO practice question is part of Courseiva's free NVIDIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the NCP-AIO exam.