Q71 — AWS SAA-C03 Ch.14

Question 71 of 100 | ← Chapter 14

Q1071. A company is developing a containerized web application that needs to be highly available and scalable.The application requires access to GPU resources.How should the company run the containerized application?

• A. Package the application as an AWS Lambda function in a container image. Use Lambda to run the containerized application on a runtime with GPU access.
• B. Deploy the application container to Amazon Elastic Kubernetes Service (Amazon EKS). Use AWS Fargate to manage compute resources and access to GPU resources.
• C. Deploy the application container to Amazon Elastic Container Registry(Amazon ECR).Use Amazon ECR to run the containerized application with an attached GPU.
• D. Run the application on Amazon EC2 instances from a GPU instance family by using Amazon Elastic Container Service (Amazon ECS) for orchestration. ✓

Correct Answer: D. Run the application on Amazon EC2 instances from a GPU instance family by using Amazon Elastic Container Service (Amazon ECS) for orchestration.

Explanation

The best solution for running a containerized web application that requires high availability, scalability, and access to GPU resources is:D. Run the application on Amazon EC2 instances from a GPU instance family by using Amazon Elastic Container Service (Amazon ECS) for orchestration.Explanation:- Amazon EC2 GPU Instances: These instances are specifically designed for workloads that require GPU resources, making them ideal for applications that need high-performance computing capabilities, such as machine learning or graphics processing.- Amazon ECS: This service provides orchestration capabilities for managing containers and can easily scale the application based on demand.Other Options:- A. Package the application as an AWS Lambda function in a container image: AWS Lambda currently does not support GPU access, so this option is not feasible for applications requiring GPU resources.- B. Deploy the application container to Amazon Elastic Kubernetes Service (Amazon EKS): While EKS can work with GPU resources, AWS Fargate does not support GPU access. You would need to run EKS on EC2 instances that have GPU capabilities, which adds complexity.- C. Deploy the application container to Amazon Elastic Container Registry (Amazon ECR): ECR is a container registry service and does not run applications. You need to use a service like ECS or EKS to deploy and manage containers.Conclusion:Option D is the most suitable choice, as it leverages EC2 GPU instances for high-performance needs while using ECS for container orchestration, ensuring high availability and scalability for the application.