As a solutions architect, your goal is to improve reliability of an application that runs on Amazon EC$2$ instances.

What would be your BEST recommendation?

Create a CloudWatch Event rule which matches a CloudWatch Alarm triggering an AWS Lambda to provision a new EC$2$ instance, if application is not reachable.

Rearchitect the application to use AWS Lambda to move to a managed service.

$($X$)$ Move to multiple smaller instances in $2$ Availability Zones and distribute requests across the instances.

Use autoscaling to increase$/$decrease the number of instances as needed.

Provision larger instance of Amazon EC$2$ in $2$ availability zones and distribute requests across the $2$ instances.