For some applications, clients at remote sites must update primary data. The easiest way to do this is for the client to connect directly to the primary data server through the network. Replication Server distributes updates from the primary to the remote sites in the usual way.
Figure 2-5 illustrates this design.
Figure 2-5: Primary data maintenance via network
This architecture uses client connections through the WAN. It is useful for applications with large amounts of data and low update rates. Updates are executed directly against the primary data so that they can be distributed to all sites that have subscriptions for the data. At remote sites, local replicated copies of the data can be used for decision-support applications to reduce the network load.
