Impact of Global Network Latency on Strong Consistency and RTO/RPO Metrics

Strong consistency requires synchronous commits across regions, meaning write latency is equal to the round-trip time between the two farthest replicas.

In a globally distributed setup, Strong consistency is bound by the speed of light. To guarantee that a read always returns the latest write, the write must be acknowledged by a majority of replicas worldwide before the client gets a success. This effectively caps your write latency at the network RTT (e.g., 200ms between London and Sydney). This drastically impacts RPO (Recovery Point Objective), which becomes 0, but negatively affects availability during network partitions.