A stateless application runs on a single Amazon EC2 instance. During business hours the instance’s CPU frequently spikes to about 90%, and users report slow response times. Which approach is the MOST operationally efficient way to improve the application's responsiveness?

The most operationally efficient way to improve responsiveness for a stateless application experiencing high CPU is to implement auto scaling. Placing the instance in an Auto Scaling group behind an Application Load Balancer (ALB) allows the system to automatically add or remove instances based on demand. A target-tracking scaling policy based on average CPU utilization ensures that new instances are launched when CPU is high, distributing the load and improving responsiveness. The ALB handles traffic distribution to the healthy instances. Enabling CloudWatch logs and alarms only notifies of the problem, it doesn't solve it. Setting up an AWS Client VPN to connect to a private IP doesn't address the high CPU on the EC2 instance; it's a connectivity solution, not a performance one. Vertically scaling the instance via a Lambda function is a temporary fix that requires downtime and doesn't handle fluctuating loads efficiently.