SAP-C02 Design for New Solutions Practice Question
A company is building a high-performance computing (HPC) cluster on AWS for genomics research. The compute nodes require low-latency inter-node communication. Which networking solution should be used?
⚠ Common exam trap
Many exam-takers confuse Enhanced Networking (ENA) with Elastic Fabric Adapter (EFA), assuming both provide similar low-latency benefits, but only EFA offers OS-bypass for HPC inter-node communication, while ENA still relies on kernel processing.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Elastic Fabric Adapter (EFA)
Elastic Fabric Adapter (EFA) is a network interface that enables HPC and machine learning applications to achieve low-latency inter-node communication by bypassing the operating system kernel and providing OS-bypass capabilities via the Libfabric library. This is essential for tightly coupled HPC workloads like genomics research, where MPI (Message Passing Interface) jobs require microsecond-level latency and high throughput between compute nodes.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Elastic Fabric Adapter (EFA)
Why this is correct
Elastic Fabric Adapter provides kernel-bypass and OS-bypass capabilities, enabling tightly coupled HPC applications to communicate with far lower latency than standard TCP/IP over Elastic Network Adapter. It satisfies the stem's low-latency inter-node requirement, supporting MPI and NCCL collective operations essential for genomics workloads distributed across many compute nodes.
- ✗
Enhanced Networking (ENA)
Why it's wrong here
Enhanced Networking with ENA raises packet-per-second performance and reduces CPU overhead, but it does not by itself deliver the kernel-bypass, RDMA-style transport that tightly coupled HPC nodes need. ENA is correct for general high-throughput EC2 workloads; EFA targets inter-node HPC messaging.
- ✗
VPC Peering
Why it's wrong here
VPC peering connects routing between two VPCs over the AWS backbone; it does not govern latency between instances inside a single cluster. It would be correct for joining separate VPC networks. Inter-node latency is addressed through placement groups and EFA rather than peering.
- ✗
AWS Direct Connect
Why it's wrong here
AWS Direct Connect provides dedicated private connectivity between on-premises networks and AWS, not inter-node traffic inside a cluster. It suits hybrid architectures needing consistent bandwidth to AWS. For low-latency node-to-node communication, the cluster placement group and EFA are the relevant mechanisms.
Go deeper
Related to this question
About these practice questions
One of 984 original SAP-C02 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This SAP-C02 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the SAP-C02 exam.