A company manages an on-premises JavaScript front-end web application. The application is hosted on two servers secured with a corporate Active Directory. The application calls a set of Java-based…

Hosting the JavaScript front-end on Amazon S3 provides a highly available, serverless, and maintenance-free solution for static web content, while Amazon Cognito identity pools enable secure user authentication and federation, meeting the AD integration and secure access requirements.

Hosting the web application on Amazon EC2, even with Auto Scaling, would still require managing and maintaining the underlying EC2 instances, which contradicts the explicit requirement to 'eliminate the need for server maintenance.'

Amazon API Gateway provides the required secure API layer for the microservices, and rehosting the Java-based microservices on AWS Fargate eliminates the need for server maintenance by providing serverless container compute, directly addressing the 'eliminate the need for server maintenance' requirement.

While Amazon Elastic Kubernetes Service (EKS) with API Gateway provides scalability, using EKS typically involves managing EC2 worker nodes (unless using Fargate profiles), which adds more operational overhead compared to AWS Fargate directly, making it less aligned with the 'eliminate server maintenance' goal.

Amazon RDS for MySQL is a managed service but is provisioned capacity and doesn't offer the same automatic scaling down to zero capacity or cost savings during idle periods as Aurora Serverless, making it less optimal for highly variable workloads compared to F.

Replatforming the clustered MySQL database to Amazon Aurora MySQL Serverless is ideal for variable workloads, as it automatically scales capacity up and down based on demand and incurs no cost when idle, perfectly aligning with the 'eliminate the need for server maintenance' and cost optimization goals for the given usage pattern.