The term active/active refers to a configuration allowing two instances of the cluster application. But only one instance of the application has data access at one time. Two disk controllers can simultaneously process I/O commands sent from one or more host computers to an array of disks. The computer systems are tasked with synchronization of the access. If one RAID controller in an active/active with failover configuration ceases to operate properly, the surviving RAID controller automatically assumes its workload. The application need only enable the data selection to "resume" activity.

The advantage of this type of configuration is that both computers are doing real work. This translates into reduced cost for the total system.

In an active/standby configuration only one computer or node is actively processing all of the users for the cluster. The other computer, in this two-node configuration, is not doing any processing work at all. It is strictly in a standby mode waiting for a failure in the primary computer. If the primary computer fails, the standby node will immediately take over processing responsibilities. Active/passive goes the active/standby one better. With active/passive, the passive server is used for additional duties such as noncritical file or application serving.

The virtual server's network name and IP address are automatically transferred from the failing server to a surviving cluster node. The process of address failover is completely transparent to client nodes and does not require any software changes or additions to...