SAA-C03 Question 469
Single answerYour company runs a web application on Amazon EC2 instances behind an Application Load Balancer. Traffic to your application is highly variable, with frequent spikes during promotional campaigns. You want to ensure the application can automatically handle traffic spikes efficiently while minimizing costs. Which scaling strategy should you choose?
- A
Set up an Auto Scaling group with dynamic scaling policies based on CPU utilization metrics.
- B
Manually add and remove EC2 instances from the Auto Scaling group during traffic spikes.
- C
Configure an Auto Scaling group with scheduled scaling policies aligned to your promotional campaign times.
- D
Enable hibernation for the EC2 instances to quickly resume operations during traffic spikes.
Show answer and explanation
Correct answer: A
Explanation
Dynamic scaling policies in an Auto Scaling group allow you to automatically adjust the number of EC2 instances based on real-time metrics, such as CPU utilization or application load. This strategy is ideal for handling highly variable traffic patterns efficiently, as it ensures the application scales out during spikes and scales in when traffic decreases, minimizing costs. Other options either require manual intervention, are designed for predictable traffic patterns, or do not address scaling at all.
- A. Correct.
This is the correct option. Dynamic scaling policies allow the Auto Scaling group to automatically scale in or out based on real-time metrics like CPU utilization. This ensures efficient handling of traffic spikes without manual intervention.
- B. Incorrect.
This is incorrect. Manually managing EC2 instances is not a scalable or cost-efficient solution, especially with highly variable and unpredictable traffic patterns.
- C. Incorrect.
This is incorrect. Scheduled scaling policies are effective for predictable traffic patterns but are not suitable for handling frequent and unpredictable traffic spikes.
- D. Incorrect.
This is incorrect. Hibernation is used to preserve the state of EC2 instances but does not help with scaling or handling traffic spikes efficiently.