You can add System Automation for Multiplatforms to an application server cluster configuration to automatically restart failed nodes.
When a failure occurs, the node agents and the application servers can be restarted on their respective nodes. They do not failover to another node. For more information, see the monitoring and restarting section of the WebSphere® Application Server Network Deployment V6: High Availability Solutions IBM Redbooks® publication.
WebSphere Application Server Network Deployment does not automatically restart failed services in the network deployment cluster. However, by adding System Automation for Multiplatforms to the configuration, the services can be configured to restart automatically when the node comes back online.
You can configure attached documents for high availability by auto-mounting a redundant shared file system on multiple application servers.