Free Amazon ANS-C01 Actual Exam Questions - Question 10 Discussion
workload. Amazon EC2 instances will require 10 Gbps flows and an aggregate throughput of up to
100 Gbps across many instances with low-latency communication.
Which architecture solution will optimize this workload?
Actually, D isn’t ideal since cluster placement groups require instances to be in the same AZ for low latency, and spreading across multiple AZs would add latency. That rules out cross-AZ setups for this HPC need.
B tbh doesn’t seem right because spread placement groups are more about fault tolerance than low latency or high throughput, which is key here. The question asks for low-latency and high aggregate throughput, so spreading nodes across subnets might increase latency and reduce performance. A cluster placement group keeps instances physically close, which aligns better with HPC needs. Plus, Elastic Fabric Adapter (EFA) is specifically designed for HPC workloads to provide that super low latency and high throughput, which option A nails.
Probably A, cluster placement and EFA are specifically made for HPC low latency.
Makes sense to go with A since cluster placement groups give low-latency and EFA supports HPC well. Other options split across AZs or VPCs might add latency or complexity. A seems cleanest here.
Option A - cluster placement group with EFA is best for low-latency HPC workloads needing enhanced networking and 10 Gbps+.