Site Planning for multi-target replication solution

In this scenario, data that is backed up from the primary site, Site A, is replicated to a server at the secondary, disaster recovery site, Site B. If a disaster occurs at Site A, such as failure of the server, you can use the server at Site B to recover systems. The directory container storage pool at site A can be repaired from the copies at either site B or site C. In addition, client systems at site A can restore data from the server at either site B or site C.

In this scenario, data that is backed up to the source replication server, Site A, is replicated to two target replication servers for disaster recovery, Site B and Site C. If a disaster occurs at Site A, such as failure of the server, you can use either server at Site B or server at Site C to recover systems. Alternatively, you can use the server at Site A to restore primary storage pool data at Site B or Site C, such as after a disk storage failure at either of the target replication servers.

In this scenario, local data at each site is backed up by the servers at both Site A and Site B. Data that is backed up from Site A is replicated to Site B, and backed-up data from Site B is replicated to Site A. If data that was backed up is lost at Site A, you can use the server at Site B to recover storage pool data to the server at Site A. If Site A is no longer available, you can recover the replicated data for Site A to a new system at Site B. You must size the server resources to ensure that either server has sufficient capacity to back up and restore all client nodes as part of your disaster recovery plan.

In this scenario, you configure remote servers that are relatively small to replicate data that is backed up to a larger server at the primary site. If bandwidth is limited, it might not be practical to restore systems to the remote sites. In this case, you might want to recover systems at the primary site before you replicate the backed-up data to the remote servers.

The network must have sufficient bandwidth for the expected data transfers between nodes, for replication, and for the cross-site restore operations that are required for disaster recovery. Before you proceed with testing replication throughput, ensure that your network can handle the replication traffic. Calculate the required network bandwidth for the steady-state requirement by applying the guidelines in Estimating network bandwidth required for replication (V7.1.1).

The network connection is often a shared resource. Plan the time of day to schedule node replication to run to avoid a conflict with other resource users. Also, network controls might limit activity to only a portion of the bandwidth. There are no controls in IBM Storage Protect to restrict network usage.

To set up the data protection solution across two sites, you must replicate data initially from Site A to the target replication server at Site B. To ensure that the initial replication is successful, you must determine whether you have the network bandwidth, processor resources, and time available to replicate the data. You might have to plan for replicating the initial full backups across several days. If you cannot extend the schedule for the initial backups, you can replicate data from Site A to Site B without using the network. For example, you can export and import the backed-up data by using media or you can temporarily locate the source and target replication servers on the same site.

For the multi-target replication solution, the daily data ingestion and total data retention must be within the capacity of the configurations. For example, a large configuration has a data ingestion capacity of up to 60 TB per day, including node replication. In cases where the backup requirements exceed the capacity of a single server, you can configure a solution that uses multiple servers to achieve the required capacity.

The server configuration must meet or exceed the requirements for the multi-target replication solution.

A data replication solution is most efficient when a single, offsite copy of the backed-up data meets your data protection and risk mitigation requirements. In this case, the single copy of the data is maintained off-site at the location of a replication server. If you require multiple backed-up data copies, you can consider the multi-target replication solution to create copies of data to two target replication servers.