Data replication for disaster recovery is distributed across three layers. Which layer is NOT part of this distribution?

• A. Host layer
• B. Network layer
• C. Storage layer
• D. Application layer

Answer: (D)

Data replication for disaster recovery is implemented across three practical layers: the host layer, the network layer, and the storage layer. The host layer involves the systems that run applications and coordinate replication to ensure client-visible consistency. The network layer handles the transmission of replication data between sites, including bandwidth, latency, and reliability concerns. The storage layer is where the actual data copies reside and are kept synchronized across storage arrays or devices, using snapshots, mirroring, or synchronous/asynchronous replication.

The application layer sits above these layers and provides the business logic and services that use the data. While applications rely on replicated data, the replication mechanics themselves are not part of this layer. Therefore, the layer not part of the replication distribution is the application layer.