Redundancy eliminates single points of failure (SPOF).

For example, a Load Balancer backup node or clustered Web and appservers.

Some components, like the Deployment Manager do not support any automatic backup/failover.

• Two Load Balancers

  The one on Server A is the primary (active) Load Balancer. It is synchronized, through a heartbeat, with a backup Load Balancer (in standby status) on another machine, Server B.

• Two Web servers

  Both of them receive requests from the Load Balancer and share the requests that come from the Internet. Each one is installed on a different machine.

• An appserver cluster

  The cluster implements vertical and horizontal scaling.

• Eight cluster members

  Two on each Application Server Node.

• Four Application Server Nodes

  Each one hosting two appservers. The nodes on Server E are independent installations. The nodes on Server F are profiles of a single installation.

• Two database servers

  Use a HA (high availability) software product. This means that one copy of the database is the one that is being used and the other one is a replica that will replace the first one if it fails.

• Two LDAP servers

  Use a HA (high availability) software product. This means that one copy of the database is the one that is being used and the other one is a replica that will replace the first one if it fails.

Next