SAP-C02 Question 451
Select 3A company is designing a high-performance computational workload that requires low-latency network communication between EC2 instances. The workload is also expected to scale dynamically based on demand. Which combination of configurations should the company implement to meet these requirements?
- A
Deploy the EC2 instances in a cluster placement group.
- B
Use Auto Scaling groups to dynamically scale the EC2 instances.
- C
Configure the EC2 instances to use Elastic Network Adapters (ENA).
- D
Launch EC2 instances in a spread placement group.
- E
Use On-Demand Instances only to respond to scaling events.
Show answer and explanation
Correct answers: A, B, C
Explanation
To meet the requirements of a high-performance computational workload with low-latency network communication and dynamic scaling, the company should use cluster placement groups to enable low-latency, high-throughput networking between instances, Auto Scaling groups to handle dynamic scaling, and Elastic Network Adapters for high-performance networking. Spread placement groups and relying only on On-Demand Instances do not fulfill the requirements for low-latency communication or efficient scaling.
- A. Correct.
Deploying EC2 instances in a cluster placement group ensures low-latency, high-throughput network communication between instances, making it suitable for high-performance computational workloads.
- B. Correct.
Using Auto Scaling groups allows the system to dynamically scale the number of EC2 instances based on demand, which is essential for handling variable workloads efficiently.
- C. Correct.
Elastic Network Adapters (ENA) are required for high-performance networking and ensure that the instances can support low-latency communication and high network throughput.
- D. Incorrect.
Spread placement groups are designed for maximizing instance isolation, not for low-latency communication, making them unsuitable for this use case.
- E. Incorrect.
Relying solely on On-Demand Instances does not ensure cost efficiency or rapid scaling, and it does not directly address the requirement for low-latency network communication.