A company runs a real-time fraud detection service that requires low-latency inference. The service runs 24/7 and has a predictable baseline load with occasional spikes. The company wants to reduce compute costs without affecting latency. Which approach is most cost-effective?

Use spot instances for the baseline load and on-demand for spikes.

Use on-demand instances for everything and rely on autoscaling.

Use a larger instance type to handle spikes without autoscaling.

Use reserved instances for the baseline load and autoscaling with on-demand for spikes.

A company has multiple AI projects running in the cloud. The finance team wants to allocate costs to each project accurately. The projects use shared storage and compute resources. What should the company implement?

Use a single budget for all projects.

Create separate cloud accounts for each project.

Use resource tags on all resources to track costs by project.

Review the bill manually and estimate costs.

A company runs a batch inference workload every night that processes customer data. The job typically takes 2 hours on a GPU instance. The data science team notices that the job finishes in 90 minutes on a larger GPU instance, but the cost per hour is 60% higher. The job is not time-sensitive and can run any time before 8 AM. What is the most cost-effective approach?

Use the larger GPU instance to reduce runtime and lower total cost.

Keep the original GPU instance and run the job on a spot instance.

Switch to a CPU-only instance and accept a longer runtime.

Run the job on a reserved instance to get a lower hourly rate.

A company stores genomic data for AI research. The data is accessed frequently during active research projects, but some projects are inactive for months. The company wants to reduce storage costs without losing the ability to quickly resume research. What is the best strategy?

Delete inactive project data to reduce costs.

Store all data in hot storage to ensure quick access.

Use lifecycle policies to move inactive project data to cool storage, and back to hot when the project becomes active.

Compress all data to reduce storage size.

A company is building an AI system that ingests data from multiple sources, processes it in real-time, and stores the results for 5 years for compliance. The data is accessed frequently for the first month, then rarely. The processing requires low-latency compute. The company wants to minimize total cost of ownership. Which combination of strategies is most effective?

Use on-demand compute and store all data in hot storage.

Use serverless compute and store data in archive storage from day one.

Use spot instances for compute and store data in cool storage.

Use reserved compute instances and lifecycle policies to tier data to archive after 30 days.

NetAppCertified AI Expert

Domain 5Objective 4

Describe Solutions to Optimize Cost CERTIFIED-AI-EXPERT Practice Questions (Page 1)

Part of the AI Common Challenges domain, which accounts for 22% of the CERTIFIED-AI-EXPERT exam. NetApp does not publish an official question count, but from its 90-minute exam (~35–60 total, ~8–13 in this domain), expect 1–2 from this objective — we provide 26 practice questions to prepare you well beyond it. (estimate)

26questions here

6free pages

5concepts

22%of the exam

Questions 1–5

1expert · hard
A company runs a real-time fraud detection service that requires low-latency inference. The service runs 24/7 and has a predictable baseline load with occasional spikes. The company wants to reduce compute costs without affecting latency. Which approach is most cost-effective?
Select an answer first
2application · medium
A company has multiple AI projects running in the cloud. The finance team wants to allocate costs to each project accurately. The projects use shared storage and compute resources. What should the company implement?
Select an answer first
3application · medium
A company runs a batch inference workload every night that processes customer data. The job typically takes 2 hours on a GPU instance. The data science team notices that the job finishes in 90 minutes on a larger GPU instance, but the cost per hour is 60% higher. The job is not time-sensitive and can run any time before 8 AM. What is the most cost-effective approach?
Select an answer first
4expert · medium
A company stores genomic data for AI research. The data is accessed frequently during active research projects, but some projects are inactive for months. The company wants to reduce storage costs without losing the ability to quickly resume research. What is the best strategy?
Select an answer first
5expert · hard
A company is building an AI system that ingests data from multiple sources, processes it in real-time, and stores the results for 5 years for compliance. The data is accessed frequently for the first month, then rarely. The processing requires low-latency compute. The company wants to minimize total cost of ownership. Which combination of strategies is most effective?
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.

Describe Solutions to Optimize Cost CERTIFIED-AI-EXPERT Practice Questions (Page 1)

Questions 1–5

A company runs a real-time fraud detection service that requires low-latency inference. The service runs 24/7 and has a predictable baseline load with occasional spikes. The company wants to reduce compute costs without affecting latency. Which approach is most cost-effective?

A company has multiple AI projects running in the cloud. The finance team wants to allocate costs to each project accurately. The projects use shared storage and compute resources. What should the company implement?

A company stores genomic data for AI research. The data is accessed frequently during active research projects, but some projects are inactive for months. The company wants to reduce storage costs without losing the ability to quickly resume research. What is the best strategy?

Keep studying — AI Common Challenges