A company runs a containerized web application on AWS Fargate. The application is memory-bound and currently uses 2 GB of memory and 1 vCPU. The company observes that the application frequently runs out of memory and the CPU utilization is consistently below 20%. The company wants to resolve the memory issue without over-provisioning CPU. What should the company do?

Move the application to an EC2 instance with 4 GB of memory and 2 vCPUs.

Increase both the task memory to 4 GB and the CPU to 2 vCPUs.

Increase the task memory to 4 GB and keep the CPU at 1 vCPU.

Increase the task CPU to 2 vCPUs and keep the memory at 2 GB.

A research team runs a machine learning training job that requires high-performance GPU compute for a few hours each week. The job is not time-sensitive and can be interrupted. The team wants to minimize compute costs. Which EC2 purchasing option and instance type should the team use?

Dedicated Host g4dn.xlarge instances

On-Demand g4dn.xlarge instances

A company has a web server that typically runs at low CPU utilization but occasionally experiences short spikes in traffic. The company wants to minimize costs while ensuring the server can handle the spikes. Which EC2 instance type is most appropriate?

Memory-optimized instances (e.g., r5)

Compute-optimized instances (e.g., c5)

A company runs a containerized application on Amazon EKS. The application has variable traffic and the company wants to automatically scale the number of pods based on CPU utilization. The company also wants to scale the number of worker nodes in the cluster based on the pending pods. What should the company do?

Use an Application Auto Scaling target tracking policy for the EKS cluster and the node group.

Use the Kubernetes Vertical Pod Autoscaler (VPA) for pod scaling and the Cluster Autoscaler for node scaling.

Use a scheduled scaling policy for pods and nodes based on known traffic patterns.

Use the Kubernetes Horizontal Pod Autoscaler (HPA) for pod scaling and the Cluster Autoscaler for node scaling.

AWSCertified Solutions Architect - Associate

Domain 3Objective 2

Task 3.2: Design High-Performing and Elastic Compute Solutions SAA-C03 Practice Questions (Page 4)

Part of the Design High-Performing Architectures domain, which makes up ~30% of our current practice bank. AWS does not publish an official question count, but from its 130-minute exam (~50–85 total, ~15–26 in this domain), expect 3–5 from this objective — we provide 30 practice questions to prepare you well beyond it. (estimate)

30questions here

6free pages

10concepts

10 Concepts in This Objective

AWS compute services overview
Distributed computing on AWS global infrastructure
Queuing and messaging patterns
Scalability capabilities and use cases
Serverless technologies and patterns
Container orchestration with ECS and EKS
Decoupling workloads for independent scaling
Metrics and conditions for scaling actions
Selecting compute options and features
Selecting resource type and size

Questions 16–20

16foundation · easy
A company wants to automatically scale its EC2 instances, DynamoDB tables, and Aurora replicas based on a unified scaling policy. Which AWS service provides a single interface to scale multiple AWS resources?
Select an answer first
17application · medium
A company runs a containerized web application on AWS Fargate. The application is memory-bound and currently uses 2 GB of memory and 1 vCPU. The company observes that the application frequently runs out of memory and the CPU utilization is consistently below 20%. The company wants to resolve the memory issue without over-provisioning CPU. What should the company do?
Select an answer first
18application · medium
A research team runs a machine learning training job that requires high-performance GPU compute for a few hours each week. The job is not time-sensitive and can be interrupted. The team wants to minimize compute costs. Which EC2 purchasing option and instance type should the team use?
Select an answer first
19foundation · easy
A company has a web server that typically runs at low CPU utilization but occasionally experiences short spikes in traffic. The company wants to minimize costs while ensuring the server can handle the spikes. Which EC2 instance type is most appropriate?
Select an answer first
20application · medium
A company runs a containerized application on Amazon EKS. The application has variable traffic and the company wants to automatically scale the number of pods based on CPU utilization. The company also wants to scale the number of worker nodes in the cluster based on the pending pods. What should the company do?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to SAA-C03

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 AWS. “SAA-C03” is a trademark of its owner, used for identification only.

Task 3.2: Design High-Performing and Elastic Compute Solutions SAA-C03 Practice Questions (Page 4)

Questions 16–20

A company wants to automatically scale its EC2 instances, DynamoDB tables, and Aurora replicas based on a unified scaling policy. Which AWS service provides a single interface to scale multiple AWS resources?

A research team runs a machine learning training job that requires high-performance GPU compute for a few hours each week. The job is not time-sensitive and can be interrupted. The team wants to minimize compute costs. Which EC2 purchasing option and instance type should the team use?

A company has a web server that typically runs at low CPU utilization but occasionally experiences short spikes in traffic. The company wants to minimize costs while ensuring the server can handle the spikes. Which EC2 instance type is most appropriate?

Keep studying — Design High-Performing Architectures