A company runs a three-tier web app on Amazon EC2 in a single Availability Zone. The app uses a self-managed MySQL database on an EC2 instance with a 1 TB Provisioned IOPS SSD (io2) EBS volume. Peak traffic is expected to require 1,000 IOPS for reads and 1,000 IOPS for writes. The company wants to minimize disruptions, stabilize performance, reduce costs, retain capacity for double the IOPS, and move the database tier to a fully managed, highly available, fault-tolerant solution. Which option meets these requirements MOST cost-effectively?

The correct option is to use a Multi-AZ deployment of an Amazon RDS for MySQL DB instance with a General Purpose SSD (gp2) EBS volume. This solution provides high availability and fault tolerance through Multi-AZ, automatically handles database management, and stabilizes performance. gp2 volumes are cost-effective for the required 2,000 total IOPS (1,000 read + 1,000 write), as gp2 provides 3 IOPS per GB, meaning a 1TB volume would deliver 3,000 IOPS, exceeding the 2,000 IOPS requirement and allowing for double the capacity. Using an io2 Block Express EBS volume with RDS would be significantly more expensive than gp2 and is overkill for the stated IOPS requirement. Amazon S3 Intelligent-Tiering is for object storage, not relational databases. Hosting the database on two large EC2 instances in active-passive mode is not a fully managed solution and would require significant operational overhead, failing to meet the "fully managed" requirement.