An application migrated to AWS has a static website frontend currently running on two EC2 instances behind an ALB and a Python backend on three EC2 instances behind another ALB. Instances are large On-Demand general-purpose types sized for on-premises peak load, but traffic is concentrated during lunchtime and light otherwise. The solutions architect must reduce infrastructure cost without harming availability. Which combination of actions should be taken? (Choose two.)

Moving the static website frontend to Amazon S3 is highly cost-effective because S3 is designed for static content hosting, eliminating the need for EC2 instances and their associated costs. It also offers high availability and scalability inherently. Migrating the backend Python application to general-purpose burstable EC2 instances (like T-series) is suitable because the traffic is concentrated during lunchtime and light otherwise. Burstable instances provide a baseline performance with the ability to burst above it when needed, which is more cost-efficient for fluctuating workloads than continuously running large On-Demand instances. Changing all EC2 instances to compute-optimized instances might not be optimal as the problem states general-purpose instances were used, and the issue is underutilization, not a lack of CPU power. Deploying the frontend to Elastic Beanstalk using the same instance type doesn't address the cost issue of over-provisioned instances. Changing all backend EC2 instances to Spot Instances could lead to availability issues if instances are terminated during peak load, which contradicts the requirement of not harming availability.