A company hosts a public web application on AWS. The website has a three-tier architecture. The frontend web tier is comprised of Amazon EC2 instances in an Auto Scaling group. The application tier is a second Auto Scaling group. The database tier is an Amazon RDS database.

The company has configured the Auto Scaling groups to handle the application's normal level of demand. During an unexpected spike in demand, the company notices a long delay in the startup time when the frontend and application layers scale out. The company needs to improve the scaling performance of the application without negatively affecting the user experience.

Which solution will meet these requirements MOST cost-effectively?