Scaling policies (target tracking, step, simple) in AWS - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ service_behavior

intermediate

How does a target tracking scaling policy behave?

Consider an AWS Auto Scaling group with a target tracking scaling policy set to maintain an average CPU utilization of 50%. What happens when the average CPU utilization rises to 70%?

A The policy increases the number of instances to bring CPU utilization back to 50%.

B The policy decreases the number of instances to reduce CPU utilization further.

C The policy does nothing because CPU utilization is above the target.

D The policy terminates all instances to reset the group.

💡 Hint

Think about how target tracking tries to keep a metric close to a set value.

Target tracking scaling policies automatically adjust the number of instances to keep the specified metric (CPU utilization) near the target value (50%). When CPU utilization rises above 50%, the policy adds instances to reduce the load per instance.

❓ Architecture

intermediate

Choosing between step scaling and simple scaling

You want to scale your application based on CPU usage spikes. You want to add more instances quickly when CPU usage exceeds 80%, and remove instances slowly when CPU usage drops below 30%. Which scaling policy fits best?

A Use a target tracking policy set to 55% CPU utilization.

B Use a simple scaling policy with the same cooldown for scaling out and in.

C Use manual scaling to add and remove instances as needed.

D Use a step scaling policy with different step adjustments for scaling out and scaling in.

💡 Hint

Consider which policy allows different responses based on how much the metric changes.

Step scaling policies allow you to define multiple thresholds with different scaling actions. This lets you add many instances quickly when CPU usage is high and remove instances more slowly when usage is low.

❓ security

advanced

Security considerations for scaling policies

Which IAM permission is required to allow an Auto Scaling group to execute scaling policies?

A cloudwatch:PutMetricAlarm

B ec2:ModifyInstanceAttribute

C autoscaling:ExecutePolicy

D iam:PassRole

💡 Hint

Think about the permission that lets Auto Scaling perform scaling actions.

The autoscaling:ExecutePolicy permission allows the Auto Scaling group to run scaling policies that add or remove instances.

🧠 Conceptual

advanced

Cooldown period effect on simple scaling

What is the effect of setting a cooldown period in a simple scaling policy?

A It prevents additional scaling actions until the cooldown period ends, allowing the system to stabilize.

B It immediately triggers multiple scaling actions to quickly adjust capacity.

C It disables scaling actions during peak hours.

D It resets the scaling policy to default settings after each action.

💡 Hint

Think about why you might want to wait before scaling again.

The cooldown period prevents the Auto Scaling group from launching or terminating additional instances until the previous scaling activity is complete and the system stabilizes.

✅ Best Practice

expert

Optimizing cost and performance with scaling policies

You manage an Auto Scaling group for a web app with variable traffic. You want to minimize cost while maintaining performance. Which combination of scaling policies is best?

A Use step scaling only with large step adjustments for all traffic changes.

B Use target tracking for average CPU utilization and step scaling for sudden traffic spikes.

C Use manual scaling to control instance count precisely.

D Use only simple scaling with a short cooldown period.

💡 Hint

Consider combining policies to handle steady and sudden changes efficiently.

Target tracking maintains steady performance by adjusting capacity automatically, while step scaling handles sudden spikes quickly. This combination balances cost and responsiveness.