How should the SysOps administrator configure the Auto Scaling group to meet these requirements?
Create a simple scaling policy with settings to make larger adjustments in capacity when the system is under heavy load.
Create a step scaling policy with settings to make larger adjustments in capacity when the system is under heavy load.
Create a target tracking scaling policy with settings to make larger adjustments in capacity when the system is under heavy load.
Use Amazon EC2 Auto Scaling lifecycle hooks. Adjust the Auto Scaling group’s maximum number of instances after every scaling event.
Explanations:
A simple scaling policy adjusts capacity by a fixed number of instances based on a single metric, such as CPU utilization. This does not allow for scaling adjustments based on varying traffic increases, nor does it cater for larger capacity adjustments during heavy loads.
A step scaling policy allows for more granular scaling decisions. It can add different amounts of capacity depending on the severity of the load. This provides the flexibility to scale aggressively for larger traffic increases and more conservatively for smaller ones, making it ideal for the situation described.
A target tracking policy adjusts capacity to maintain a specified target, such as average CPU utilization. While useful for maintaining stability, it does not provide the ability to scale differently based on the severity of traffic surges, as required by the scenario.
Lifecycle hooks manage the states of EC2 instances during scaling operations but do not directly influence the speed or scale of scaling adjustments. Adjusting the maximum number of instances after every scaling event would not solve the problem of timely scaling during traffic surges.
I classify that the answer is:
Create a step scaling policy with settings to make larger adjustments in capacity when the system is under heavy load.