SAP-C02 Question 450
Select 3A company has deployed a high-performance computing (HPC) application that requires low-latency network communication between EC2 instances. The workload also experiences unpredictable traffic spikes, and the company wants to ensure scalability while maintaining performance. Which combination of AWS features will best meet these requirements?
- A
Launch the EC2 instances within a cluster placement group
- B
Use Auto Scaling groups to manage the fleet of EC2 instances
- C
Deploy EC2 instances in a spread placement group
- D
Enable Elastic Fabric Adapter (EFA) on the EC2 instances
- E
Use Spot Instances to reduce costs for the workload
Show answer and explanation
Correct answers: A, B, D
Explanation
For high-performance computing workloads requiring low-latency network communication, using a cluster placement group and enabling Elastic Fabric Adapter (EFA) ensures optimal network performance. Auto Scaling groups complement this architecture by dynamically scaling the number of EC2 instances to handle traffic spikes. A spread placement group and Spot Instances, while useful in other scenarios, do not meet the specific requirements of low-latency, high-performance workloads.
- A. Correct.
Correct. A cluster placement group ensures low-latency, high-bandwidth communication between EC2 instances, which is critical for high-performance computing workloads.
- B. Correct.
Correct. Auto Scaling groups allow the environment to scale in response to traffic spikes, ensuring the application can handle unpredictable workloads while maintaining performance.
- C. Incorrect.
Incorrect. A spread placement group is used to distribute instances across underlying hardware for fault tolerance, not for low-latency or high-performance computing requirements.
- D. Correct.
Correct. Elastic Fabric Adapter (EFA) provides low-latency, high-throughput networking, which is ideal for high-performance computing applications.
- E. Incorrect.
Incorrect. While Spot Instances can reduce costs, they do not guarantee the low latency or consistent performance required for high-performance computing workloads.