SOA-C02 Question 321
Single answerYour application running on an Amazon EC2 instance is experiencing performance issues during peak traffic hours. Amazon CloudWatch metrics indicate that the CPU utilization of the instance frequently exceeds 90%, and the application response time significantly increases. Which of the following actions should you take to address the performance issue while optimizing costs?
- A
Upgrade the current instance type to one with more CPU and memory capacity.
- B
Enable Auto Scaling and configure it to launch additional instances when CPU utilization exceeds 85%.
- C
Move the application to an Amazon ECS cluster for better resource management.
- D
Switch the instance to a burstable instance type (T-series) to benefit from CPU credits.
Show answer and explanation
Correct answer: A
Explanation
The scenario indicates high CPU utilization as the primary issue. Upgrading the instance type to one with greater compute capacity addresses the bottleneck directly, ensuring the application has sufficient resources to handle peak traffic. While other options may be valid under different circumstances, they are either less effective or irrelevant for solving the specific problem described.
- A. Correct.
Upgrading the instance type to one with higher CPU and memory capacity is the most direct action to address high CPU utilization and improve application performance. This option provides immediate relief to the performance bottleneck.
- B. Incorrect.
While enabling Auto Scaling can be a good solution for horizontal scaling, it may not be cost-effective for an application experiencing high CPU usage on a single instance. Additionally, it requires the application to be architected for distributed scaling, which is not mentioned in the scenario.
- C. Incorrect.
Moving the application to an Amazon ECS cluster may improve resource management, but it is not directly relevant to the immediate issue of high CPU utilization on the existing EC2 instance. This solution would involve significant re-architecture and migration effort.
- D. Incorrect.
Switching to a burstable instance type (T-series) is not suitable for applications with sustained high CPU utilization, as these instance types are designed for workloads with intermittent bursts of activity. The performance issue would persist due to the lack of sustained compute capacity.