A company runs a batch AI training job that is CPU-intensive and can be interrupted. They want to use the spare capacity of the cluster without affecting other critical workloads. Which Kubernetes feature should they use?

A StatefulSet with persistent storage.

A Deployment with a high replica count.

A PriorityClass with a low priority value.

A DaemonSet running on all nodes.

A company runs a real-time AI inference service that experiences unpredictable traffic spikes. The service is CPU-bound. They want to minimize cost while ensuring low latency during spikes. They are considering using a HorizontalPodAutoscaler (HPA) with a target CPU utilization of 60%. Which additional configuration would best help achieve their goal?

Set minReplicas to 3 and maxReplicas to 10.

Use a VerticalPodAutoscaler instead of an HPA.

Set the HPA to scale based on memory utilization instead of CPU.

Set minReplicas to 1 and maxReplicas to 10.

NetAppCertified AI Expert

Domain 3Objective 4

Demonstrate the Ability to Execute AI Workloads at Scale with Kubernetes (Trident) CERTIFIED-AI-EXPERT Practice Questions (Page 5)

Part of the AI Software Architectures domain, which accounts for 18% of the CERTIFIED-AI-EXPERT exam. NetApp does not publish an official question count, but from its 90-minute exam (~35–60 total, ~6–11 in this domain), expect 1–2 from this objective — we provide 36 practice questions to prepare you well beyond it. (estimate)

36questions here

8free pages

9concepts

18%of the exam

Questions 21–25

21application · medium
A company runs a batch AI training job that is CPU-intensive and can be interrupted. They want to use the spare capacity of the cluster without affecting other critical workloads. Which Kubernetes feature should they use?
Select an answer first
22application · easy
A team wants to deploy a multi-container AI application where a front-end service needs to communicate with a back-end model-serving pod. They want to expose the back-end service only within the cluster, not to the internet. What should they use?
Select an answer first
23foundation · easy
Which Kubernetes object can automatically adjust the number of pod replicas based on observed CPU utilization or custom metrics?
Select an answer first
24foundation · easy
An AI training pod is stuck in the Pending state. Which command would you use to see detailed information about the pod, including scheduling events and any resource constraints?
Select an answer first
25expert · hard
A company runs a real-time AI inference service that experiences unpredictable traffic spikes. The service is CPU-bound. They want to minimize cost while ensuring low latency during spikes. They are considering using a HorizontalPodAutoscaler (HPA) with a target CPU utilization of 60%. Which additional configuration would best help achieve their goal?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to CERTIFIED-AI-EXPERT

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by NetApp. “CERTIFIED-AI-EXPERT” is a trademark of its owner, used for identification only.

Demonstrate the Ability to Execute AI Workloads at Scale with Kubernetes (Trident) CERTIFIED-AI-EXPERT Practice Questions (Page 5)

Questions 21–25

A company runs a batch AI training job that is CPU-intensive and can be interrupted. They want to use the spare capacity of the cluster without affecting other critical workloads. Which Kubernetes feature should they use?

A team wants to deploy a multi-container AI application where a front-end service needs to communicate with a back-end model-serving pod. They want to expose the back-end service only within the cluster, not to the internet. What should they use?

Which Kubernetes object can automatically adjust the number of pod replicas based on observed CPU utilization or custom metrics?

An AI training pod is stuck in the Pending state. Which command would you use to see detailed information about the pod, including scheduling events and any resource constraints?

Keep studying — AI Software Architectures