To enable true resilience for our application, the standard approach to handling disaster recovery is to set up a duplicate instance in another data center. Typically, these should be sufficiently apart to ensure that if there is a local event significant in stature to impact our standard operations, then at least one of our data centers is not affected. Irrespective of our requirements of recovery time and recovery points, we will need a process for migrating data between our sites; therefore, if there is a need to failover to an alternative site, we will have the required data there ready for us.

We can configure Hazelcast to push entries from our local cluster to a remote cluster by defining a Wide Area Network (WAN) replication policy. So, we add a policy to push changes to Paris from our London configuration.

<hazelcast>
  <group>
    <name>london</name>
    <password>lon-pass</password>
  </group>

  <wan-replication...