Hard1 markMultiple Choice
Domain 3.2: Compute PerformancePerformanceEC2HPCNetworking

AWS SAA-C03 · Question 36 · Domain 3.2: Compute Performance

A company is deploying a high-performance computing (HPC) cluster on AWS. The cluster consists of tightly coupled Amazon EC2 instances that require the lowest possible network latency and highest network throughput between them. <br/><br/>Which TWO configurations should the solutions architect apply? (Select TWO.)

Answer options:

A.

Launch the instances in a Cluster Placement Group.

B.

Launch the instances in a Spread Placement Group.

C.

Enable Elastic Fabric Adapter (EFA) on the instances.

D.

Use AWS Global Accelerator.

E.

Attach multiple Elastic Network Interfaces (ENIs) to each instance.

How to approach this question

Identify the placement group and network interface designed specifically for HPC.

Full Answer

Launch the instances in a Cluster Placement Group.<br/>Enable Elastic Fabric Adapter (EFA) on the instances.
For tightly coupled HPC applications, you should use a Cluster Placement Group to ensure instances are physically close to each other in the data center. Additionally, using an Elastic Fabric Adapter (EFA) provides OS bypass capabilities, significantly reducing network latency between instances.

Common mistakes

Selecting Spread Placement Group.

Practice the full AWS SAA-C03 Practice Exam 4

65 questions · hints · full answers · grading

More questions from this exam