IBM Db2 Mirror for IBM i functionality
Critical business applications require a high availability of the underlying databases. With IBM Db2 Mirror for IBM i, there is a solution to keep two Db2 databases in the same data center synchronized so that if one database goes down, the other database will be accessed without any user interaction. In this case, the critical factor of the Recovery Time Objective (RTO) is zero and could not be better.
Restrictions
Customers need to be aware of several restrictions on IBM Db2 Mirror. For synchronous replication, a high-speed connection between the two servers is required, which implies that they need to be in the same data center. There are other restrictions, like not all objects being replicated and integrated file system (IFS) is not supported, which conclude that this solution is not suitable for a disaster recovery solution; even IBM recommends adding separate HA/DR solutions for this use case.
Rocket iCluster Provides
Rocket iCluster provides everything a customer needs to cover any HA/DR scenario: It creates exact copies of IBM i nodes, including objects that IBM Db2 Mirror does not support for i, IFS support, and long-distance replication to any data center on-premises or in the cloud. High-performant sync checks ensure the backup holds up-to-date data before switching over.
IBM Db2 Mirror for IBM i and Rocket iCluster 9.1.0 - Working together
Rocket Software is happy to announce that with Rocket iCluster 9.1.0, we support the use of IBM Db2 Mirror for i together with Rocket iCluster on the same nodes. Customers can now benefit from a zero RTO and workload balancing with IBM Db2 Mirror for i and the confidence that in case of a local disaster in the data center, there is still a backup node available in another location that can be up and running in minutes.
We recommend refraining from replicating the same source and target nodes with IBM Db2 Mirror for IBM i and Rocket iCluster, as this could result in replication conflicts and is not an everyday use case. Choosing different target nodes is always the best topology for high availability and disaster recovery.
#iCluster #IBMi------------------------------
Harald Braeunlich
Senior Product Manager
------------------------------