Q23Network Design
A network engineer must design an architecture for a high-performance computing (HPC) workload. Amazon EC2 instances will need 10 Gbps flows and aggregate throughput of up to 100 Gbps across numerous instances, with low-latency communication. Which architecture solution will optimize this workload?
← → navigate · a answer
Community votes
Discussion · 4
A 6
Option A is the most suitable solution for the high-performance computing workload:
Cluster Placement Group: A cluster placement group ensures that EC2 instances are placed close together within the same Availability Zone to minimize network latency and provide high-bandwidth, low-latency communication between instances. This is crucial for an HPC workload where low-latency communication is essential.
Single Subnet of a VPC: Placing nodes in a single subnet of a VPC ensures that the EC2 instances are closely located within the same subnet, further reducing network latency.
Elastic Fabric Adapter (EFA): EFA is a network interface designed specifically for HPC workloads that require low-latency and high-bandwidth communication. Using EFA drivers on EC2 instances ensures optimal performance for communication between instances.
A 3
A indeed correct
A 2
A is correct
1
A
https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/placement-groups.html#placement-groups-limitations-spread