What factors influence cross-region replication latency and its impact on RPO?

What factors influence cross-region replication latency and its impact on RPO?

• A. CPU speed and RAM size only.
• B. Disk type only.
• C. Network bandwidth, distance, replication method, and workload.
• D. User login patterns.

Answer: (C)

Cross-region replication latency is shaped by how quickly data can be moved across the network and how the replication is implemented. The factors that matter are network bandwidth, distance, replication method, and workload. Bandwidth sets the maximum data transfer rate; greater distance adds propagation delay and can introduce variability due to routing and congestion. The replication method determines when data is considered committed on the remote site and how much lag is tolerable (for example, synchronous replication aims for near-zero lag but can be limited by network speed, while asynchronous replication allows more lag). Workload captures the rate and size of changes being replicated; a higher change rate or bursty workload creates more data to replicate and can widen the lag if the destination can’t keep up. These elements together influence the Recovery Point Objective, which represents the maximum tolerable data loss in an outage. If replication lags, the RPO grows because the replica may not reflect the most recent transactions. In contrast, factors like CPU speed, RAM size, or disk type mainly affect performance at the local primary system rather than the inherent cross-region replication delay, and user login patterns don’t drive the replication lag between sites.

Cross-region replication latency is shaped by how quickly data can be moved across the network and how the replication is implemented. The factors that matter are network bandwidth, distance, replication method, and workload. Bandwidth sets the maximum data transfer rate; greater distance adds propagation delay and can introduce variability due to routing and congestion. The replication method determines when data is considered committed on the remote site and how much lag is tolerable (for example, synchronous replication aims for near-zero lag but can be limited by network speed, while asynchronous replication allows more lag). Workload captures the rate and size of changes being replicated; a higher change rate or bursty workload creates more data to replicate and can widen the lag if the destination can’t keep up.

These elements together influence the Recovery Point Objective, which represents the maximum tolerable data loss in an outage. If replication lags, the RPO grows because the replica may not reflect the most recent transactions. In contrast, factors like CPU speed, RAM size, or disk type mainly affect performance at the local primary system rather than the inherent cross-region replication delay, and user login patterns don’t drive the replication lag between sites.