Availability is the probability that a system, at a point in time, will be operational and able to deliver the requested services.
The availability requirement of a P2P system, like any other system, is dependent upon the use for which the system is deployed. Traditionally, high availability requirements can usually be catered for by using replication or by employing fast methods of re-initialising the services that become unavailable.
Replication can, theoretically, be utilised by a P2P system because the systems very nature can actually support it, in fact replication is almost a by-product of such a system and as such could be exploited for availability purposes. Re-initialising a P2P system on the other hand could be more of a problem, especially for distributed applications where only some of the nodes are re-initialised.