How about IPAT via IP aliases and IPAT via IP replacements?

When IPAT via IP Aliases is used

The service IP label/address is aliased onto the same network interface as an existing communications interface. 

That is, multiple IP addresses/labels are configured on the same network interface at the same time. In this configuration, all IP addresses/labels that you define must be configured on different subnets.

This method can save hardware, but requires additional subnets.

When IPAT via IP Replacement is used

The service IP label/address replaces the existing IP label/address on the network interface. 

That is, only one IP label/address is configured on the same network interface at the same time. In this configuration, two IP addresses/labels on a node can share a subnet, while a backup IP label/address on the node must be on a different subnet. 

This method can save subnets but requires additional hardware.

-----------------------------------------------------------------

Notes:

IPAT via IP Aliases provides the following advantages over the IPAT via IP Replacement scheme:

**Running IP Address Takeover via IP Aliases is faster than running IPAT via IP Replacement, because moving the IP address and the hardware address takes considerably longer than simply moving the IP

address.

**IP aliasing allows co-existence of multiple service labels on the same network interface— you can use fewer physical network interface cards in your cluster. Note that upon fallover, HACMP equally

distributes aliases between available network interface cards.