Target Tracking Scaling

5 minutes 5 Questions

Target Tracking Scaling is a dynamic scaling policy mechanism that enables you to define a target value for specific CloudWatch metrics, and have the scaling group automatically adjust its capacity in real-time to maintain the desired metric value. This can be useful for scenarios where your application needs to maintain a consistent level of resource utilization, response time, or throughput. Available metrics include CPU utilization, request count per target, and network bandwidth. The benefit of Target Tracking Scaling is that it simplifies the scaling process by reducing the need to manually configure and manage individual scaling policies, while still ensuring optimal performance and cost efficiency for your application.

Guide to Target Tracking Scaling

What is Target Tracking Scaling?
Target Tracking Scaling is a feature provided by AWS Auto Scaling, aimed at simplifying the set-up and maintenance of application scalability. It allows you to define a scaling strategy aimed at maintaining a specific CloudWatch metric at or close to your target value.

Why is it important?
It is important as it helps automatically adjust the application in response to demand fluctuations. It simplifies capacity provisioning and improves cost-effectiveness by increasing capacity during demand spikes and reducing capacity during demand lulls.

How does it work?
Target Tracking Scaling works by defining target tracking scaling policies, that tell Amazon EC2 Auto Scaling to track a specific CloudWatch metric and adjust the number of instances as necessary to keep that metric at your chosen target value. It creates and manages Cloudwatch alarms for triggering the scaling process.

Exam Tips: Answering Questions on Target Tracking Scaling
1. Understand that Target Tracking Scaling policy maintains a specified metric at a target value, it dynamically calculates the number of instances required.
2. Be clear that when a target tracking policy is added, it might override any existing step and simple scaling policies.
3. Remember, it works only with scalable resources supporting the specified metric.
4. Be familiar with Amazon EC2 Auto Scaling and CloudWatch, as the understanding of these services is crucial for answering related questions.

Test mode:

Exam (Timed)

Practice (With explanations)

Start practice test

Question 1

When configuring a Target Scaling policy for an Auto Scaling group, what metric can you use besides CPU Utilization?

EBS Volume Size AWS Account Balance RequestCountPerTarget EC2 Instance Count

Correct Answer: RequestCountPerTarget

RequestCountPerTarget is a valid scaling policy metric. Other options such as EC2 Instance Count, EBS Volume Size, AWS Account Balance, Internet Data Transfer Rate, and Total Disk I/O Rate are either not valid scaling policy metrics or not directly related to Target Tracking Scaling policies.

Question 2

You set up a Target Tracking Scaling policy for a new application with a target value of 40% average CPU utilization. However, your instances are running at 80% CPU utilization. What should you adjust in order to reduce CPU utilization in this situation?

Set the minimum size of the Auto Scaling group to a higher value. Decrease the target value. Increase the target value. Set the maximum size of the Auto Scaling group to a lower value.

Correct Answer: Increase the target value.

Increasing the target value for CPU utilization will force the Target Tracking Scaling policy to add more instances and thus reducing the overall CPU utilization.

Question 3

You are setting up a Target Tracking Scaling policy for an Auto Scaling group to maintain an average CPU utilization of 70%. During peak load, you notice that the desired capacity never exceeds 5 instances. Which possible issue is the case?

The maximum size of the Auto Scaling group is set to 5. The refresh period is too long. The Target Tracking Scaling policy needs to be updated. The target value is too high.

Correct Answer: The maximum size of the Auto Scaling group is set to 5.

If the desired capacity never exceeds 5 instances during peak load, the maximum size of the Auto Scaling group is likely set to 5. Other factors such as target value, scaling policy or refresh period would not limit capacity.

Go Premium