Which EC2 Configuration Meets Tightly Coupled HPC Network Requirements?

Why the Answer Is Correct

A cluster placement group packs instances close together inside a single Availability Zone so they sit on the same high-bisection-bandwidth network segment, which is precisely what tightly coupled HPC node-to-node traffic needs. AWS documentation states that cluster placement groups provide the lowest latency and highest packet-per-second network performance available in EC2, with supported instances reaching very high single-flow throughput, and Elastic Fabric Adapter can be layered on top for even lower-latency HPC interconnects. The question's wording — 'low-latency network performance and high network throughput with tightly coupled node-to-node communication' — is the canonical AWS phrasing for cluster placement groups. Launching the EC2 instances into a cluster placement group therefore directly satisfies every stated requirement. Option A is the answer the questions is looking for and matches the consensus of every commenter who voted.

Why the Other Options Are Wrong

Dedicated Instance tenancy (B) changes the hardware isolation model, not the network fabric; a Dedicated Instance still communicates over the same paths as a shared-tenancy instance and offers no low-latency inter-node guarantee. Spot Instances (C) are a purchasing option: they can be interrupted with two minutes' notice and offer no placement or latency benefit, which is exactly the wrong foundation for a tightly coupled job that must keep all nodes alive simultaneously. An On-Demand Capacity Reservation (D) guarantees that capacity is available in a specific Availability Zone when you need it, which solves a capacity problem rather than a performance problem — and it can even be combined with a cluster placement group rather than replacing it. None of B, C, or D changes the topology or the physical proximity of the nodes.

Community Comment Notes

Scheldon admitted thinking the answer was D until researching further, then settled on A and linked the AWS placement groups documentation, which is the decisive source here. zinabu isolated the exact trigger phrase, noting that "tightly coupled node-to-node communication" points to a cluster placement group, and KennethNg923 reasoned the same way from the identical wording. Awsbeginner87 and Tanidanindo summed it up as everything in the scenario pointing to a cluster placement group. The lone dissent, victor78's "I THINK IT'S C,D", pairs Spot Instances with a capacity reservation — a self-defeating combination for a job that needs uninterrupted, low-latency interconnect.

Exam Strategy

When an SAA-C03 scenario mentions HPC, tightly coupled nodes, low latency, or high network throughput, scan first for a placement group answer before evaluating capacity or pricing options. Separate the verb in the requirement: 'ensure capacity' maps to capacity reservations or Reserved Instances, while 'reduce latency / increase throughput' maps to placement groups, enhanced networking, or EFA. Also remember that a cluster placement group lives inside one Availability Zone, so an option that promises multi-AZ resilience is not answering the latency requirement.