SOA-C02 Question 60
Select 3Your company runs a web application hosted on Amazon EC2 instances behind an Application Load Balancer. You have been tasked with creating an Auto Scaling plan to handle variable traffic patterns efficiently while minimizing costs. Which of the following configurations can help achieve this goal?
- A
Configure a target tracking scaling policy that adjusts the desired capacity to maintain CPU utilization at 50%
- B
Set up a scheduled scaling policy to increase the instance count by 2 every day at 9:00 AM and decrease it by 2 every night at 9:00 PM
- C
Use a step scaling policy to increase capacity by 1 instance when the average network traffic exceeds 500 Mbps
- D
Enable predictive scaling to forecast load and automatically adjust the desired capacity based on historical traffic patterns
- E
Manually adjust the desired capacity of the Auto Scaling group based on weekly traffic estimates
Show answer and explanation
Correct answers: A, B, D
Explanation
To handle variable traffic patterns efficiently while minimizing costs, a combination of target tracking policies, scheduled scaling for predictable traffic patterns, and predictive scaling for historical trend-based adjustments can be used. These approaches help ensure that the application scales dynamically and automatically, avoiding underutilization and overprovisioning. Manual adjustments and reactive-only policies may not align with the requirement for an efficient and cost-effective Auto Scaling plan.
- A. Correct.
Correct. A target tracking scaling policy dynamically adjusts capacity to maintain CPU utilization at 50%, ensuring the application can handle traffic spikes efficiently while minimizing underutilization.
- B. Correct.
Correct. Scheduled scaling can help accommodate predictable traffic patterns, such as daily peaks, by increasing or decreasing capacity at specific times.
- C. Incorrect.
Incorrect. While step scaling policies are useful, the given configuration does not align with minimizing costs or efficiently handling variable traffic patterns, as it reacts only to network traffic and may not account for other metrics like CPU utilization or load balancer request counts.
- D. Correct.
Correct. Predictive scaling uses machine learning to forecast future demand and adjusts capacity accordingly, which can efficiently handle variable traffic patterns while optimizing costs.
- E. Incorrect.
Incorrect. Manually adjusting the desired capacity is error-prone and does not align with the goal of creating an automated and efficient Auto Scaling plan.