Planning for node replication

Use a checklist of planning considerations to streamline implementation. Planning considerations include items such as the types of data that you want to replicate and the replication rules that you want to apply.

• Identify the client nodes and, if necessary, the files spaces that belong to those nodes, that have data to be replicated.
• Identify the types of data that you want to replicate. With Tivoli® Storage Manager, you can replicate backup data, archive data, and space-managed data. If you decide to replicate backup data, you must decide whether to replicate only active backup data or to replicate active and inactive backup data. In environments with large amounts of backup data, it might make sense to replicate only active backup data. The type of data that you replicate can vary by node or by file space.
• Determine the order in which data is to be replicated. If you want to replicate some data ahead of other data in a replication process, you can assign a high-priority replication rule to the data that you want to replicate first. A high-priority classification does not increase the speed of replication.
• Set up a default replication configuration. In the default configuration, all backup, archive, and space-managed data is replicated in all file spaces in all replication-enabled client nodes. You can also change the replication rules at the file space, client node, and server levels. In general, you can apply different replication rules at different levels and achieve the same results.
• Identify source and target replication servers. If you have multiple source replication servers, consider whether you want to replicate data to one or more target replication servers. A single target replication server can have multiple source replication servers. However, a source replication server can replicate data to only one target replication server at a time.
• Consider using the enterprise configuration functions of Tivoli Storage Manager to set up, manage, and synchronize policy configurations on the source and the target replication servers. For example, you can change a policy configuration on a source replication server and have the changes that are automatically distributed to the target replication server.
• Consider how you want to replicate data. You can replicate one or more client nodes or defined groups of client nodes in a single process. You can schedule replication processes at different times to control network throughput.
• Decide whether you want to implement replication gradually or at one time. You can start with a simple configuration that comprises a single source replication server and a single target replication server. After you experiment with that configuration, you can build a broader network of replicating servers.
• If you use an LDAP directory server to authenticate passwords, any target replication servers must be configured for LDAP passwords. Data that is replicated from a node that authenticates with an LDAP directory server is inaccessible if the target replication server is not configured. If your target replication server is not configured, replicated data from an LDAP node can still be written to the replication server. However, the target replication server must be configured to use an LDAP directory so that you can access the data.
• Ensure that you allocate extra space for the active log and the archive log during replication processing.
• Remember that the Tivoli Storage Manager for Space Management and HSM for Windows clients do not automatically fail over to a secondary target server. You must manually edit the dsm.sys file to connect to the secondary server. Any secondary server information in the replservername stanza and myreplicationserver option is ignored by the Tivoli Storage Manager for Space Management and HSM for Windows clients.