A county emergency-management office runs tightly coupled flood-inundation batch models on EC2. Jobs often place workers far apart across AZs, and capacity is pieced together with ad-hoc On-Demand launches that miss peak storm windows. Which continuous-improvement change best raises throughput for this HPC-style civic workload?
Select an answer to reveal the explanation.
Short Explanation
Flood models hate chatty neighbors across town—those workers need to sit in the same tightly packed neighborhood on the AWS campus. Pair a placement group with a fleet that can actually fill the seats when the storm clock is ticking.
Full Explanation
Tightly coupled batch and HPC-style flood models benefit from placement groups that keep instances close for low latency and high bisection bandwidth, combined with EC2 Fleet or Spot Fleet strategies that assemble the required capacity mix. Spreading the same tightly coupled job across Regions or tiny single instances undermines network locality and throughput. Queues help decoupling for many architectures but do not replace placement and fleet capacity for synchronized model workers.