What should the solutions architect recommend to maximize utilization while ensuring the application remains fault tolerant?
Remove some EC2 instances to increase the utilization of remaining instances.
Increase the Amazon Elastic Block Store (Amazon EBS) capacity of instances with less CPU utilization.
Modify the Auto Scaling group scaling policy to scale in and out based on a higher CPU utilization metric.
Create a new launch configuration that uses smaller instance types. Update the existing Auto Scaling group.
Explanations:
Removing EC2 instances would reduce the fault tolerance of the application, as fewer instances would be available to handle potential failures, even though CPU utilization is low. This doesn’t maintain application resilience.
Increasing the EBS capacity does not address the issue of low CPU utilization. EBS capacity is unrelated to CPU utilization and doesn’t help in maximizing resource usage or improving the fault tolerance.
Modifying the scaling policy based on higher CPU utilization is not ideal, as CPU utilization is already low. This would cause scaling actions to be triggered at suboptimal times and would not maximize the resource utilization effectively.
Using smaller instance types can increase the utilization of CPU resources, as the instances will be more likely to reach higher utilization. This helps in better resource utilization while ensuring fault tolerance by maintaining the same number of instances across Availability Zones.
As far as I’m aware, the answer is:
Create a new launch configuration that uses smaller instance types. Update the existing Auto Scaling group.