A simulation places tightly coupled compute nodes in a single Availability Zone and needs the lowest latency between those nodes. Which placement group should the nodes use?
Select an answer to reveal the explanation.
Short Explanation
You want these machines in the same aisle, not in separate buildings. A cluster placement group packs the nodes close together inside one Availability Zone, which is the low-latency layout this simulation needs. A spread group or a partition group exists to keep instances apart, and a cluster group does not stretch across zones.
Full Explanation
A cluster placement group places instances close together inside one Availability Zone so they get the lowest latency and the highest packet rate between them. The simulation is already in a single zone and wants that proximity, so the cluster group is the match. A spread placement group does the opposite: it puts each instance on distinct hardware so they do not share a host, which adds separation rather than closeness. A partition placement group divides instances into partitions so a rack-level failure does not take every node. That is a survival layout, not the lowest-latency pack. A group that places the nodes across several zones keeps them far apart. A cluster group cannot span zones, and cross-zone distance works against the latency goal. Caveat: a cluster group is one zone, so an Availability Zone failure takes the whole group; that is accepted here because the requirement is latency between the nodes. Operational check: launch the nodes together into the cluster group and confirm they land in the same zone and report placement in that group.