A company wants to migrate an existing application from its data center to the AWS Cloud to make it highly available and resilient. The application recently lost data when a database server crashed during a power outage. The company requires a solution that avoids single points of failure and can scale to meet user demand. Which solution meets these requirements?

The correct solution uses an Auto Scaling group across multiple Availability Zones (AZs) for EC2 instances, ensuring high availability and resilience for the application servers by distributing them and automatically replacing unhealthy instances. An Amazon RDS DB instance in a Multi-AZ configuration provides synchronous replication to a standby instance in a different AZ, offering automatic failover and data durability in case of a primary database failure, directly addressing the past data loss issue and avoiding a single point of failure. Incorrect options: Deploying application servers in a single AZ creates a single point of failure for the entire application tier. Deploying the database on a single EC2 instance is not highly available and does not provide automatic failover or data durability like RDS Multi-AZ. EC2 Auto Recovery only restarts an instance on the same host, not across AZs. An Amazon RDS read replica in a single AZ does not provide automatic failover for the primary database across AZs. Manual promotion is required, leading to downtime, and the read replica itself might be in the same AZ as the primary, offering no cross-AZ resilience. Using Amazon EBS Multi-Attach for shared storage between database servers on EC2 instances across multiple AZs is not a standard or recommended highly available database solution. EBS Multi-Attach is for block