SOA-C02 Question 296
Select 3A company is running multiple EC2 instances in an Auto Scaling group to handle variable web traffic. They have noticed that during off-peak hours, the instances remain underutilized, leading to unnecessary costs. Which strategies can the SysOps Administrator implement to optimize costs while maintaining application availability?
- A
Configure a scheduled scaling policy to scale down the number of instances during off-peak hours.
- B
Switch the instance types to On-Demand Instances for predictable performance and cost savings.
- C
Use EC2 Spot Instances for non-critical workloads within the Auto Scaling group.
- D
Enable AWS Cost Explorer to identify underutilized instances and take appropriate actions.
- E
Migrate to Reserved Instances for the Auto Scaling group to reduce costs.
Show answer and explanation
Correct answers: A, C, D
Explanation
To optimize costs in a dynamic environment like an Auto Scaling group, leveraging scheduled scaling to match demand, using cost-effective Spot Instances for non-critical workloads, and utilizing AWS Cost Explorer to identify underutilized resources are effective strategies. Reserved Instances are not ideal for variable workloads, and On-Demand Instances offer no additional cost savings compared to other options.
- A. Correct.
Correct: Scheduled scaling allows the Auto Scaling group to scale down during off-peak hours and scale up during peak hours, optimizing costs while maintaining availability.
- B. Incorrect.
Incorrect: On-Demand Instances are more expensive than other purchasing options, such as Spot Instances or Reserved Instances, and do not directly optimize costs for this use case.
- C. Correct.
Correct: EC2 Spot Instances are highly cost-effective and can be used for non-critical workloads, reducing overall costs in the Auto Scaling group.
- D. Correct.
Correct: AWS Cost Explorer helps identify underutilized resources, allowing the administrator to take actions to optimize costs.
- E. Incorrect.
Incorrect: Reserved Instances are better suited for predictable, steady-state workloads rather than variable workloads managed by an Auto Scaling group.