Deploy an EC2 Auto Scaling group across at least two Availability Zones and place it behind an Application Load Balancer and Add private subnets in another Availability Zone and migrate the database to Amazon RDS with Multi-AZ enabled are correct because they provide AZ-level redundancy for both the web tier and the database tier.

The first correct choice creates an Auto Scaling group that spans at least two Availability Zones and places instances behind an Application Load Balancer. This spreads web server capacity across AZs and lets the load balancer route traffic away from a failed AZ, so the web tier remains available when one AZ has an outage.

The second correct choice adds subnets in another AZ and migrates the database to Amazon RDS with Multi-AZ. RDS Multi-AZ provides synchronous replication and automatic failover to a standby in another AZ, so the database can fail over without manual intervention and the application can continue operating.

AWS Global Accelerator is not sufficient because it optimizes global network performance and front-ends public endpoints, but it does not make internal private tier deployments across AZs resilient by itself.

Create new private subnets in a second Availability Zone and host the database on Amazon EC2 in only one AZ does not provide database failover because the database remains a single AZ EC2 deployment, so an AZ failure would still disrupt the database.

Attach the current web instances to an Auto Scaling group behind an Application Load Balancer but keep them in the existing Availability Zone still concentrates all web capacity in one AZ, so the web tier would not survive an AZ outage even though it is managed by Auto Scaling and an ALB.

Full AWS Practitioner Certification Question