A company is setting up a distributed in-memory database on Amazon EC2 instances, comprising a primary node and eight worker nodes. The primary node's roles include monitoring cluster health, handling user requests, dispatching these requests to the worker nodes, and consolidating responses for clients. The worker nodes interact to replicate data partitions. To maximize performance, the company demands the least possible networking latency. Which configuration will fulfill this requirement?