What is a high availability VPN?

Essentially, a highly available VPN is one where the VPN can quickly and easily recover from component failure, and continue to provide service. Of course, high availability is almost never a black or white sort of thing; there are always degrees of availability. How long does the VPN take to recover after a component has failed? What performance impact does the insertion of high availability intelligence introduce? What is the visible impact upon a connected user after a failover has taken place?

Since performance is also a factor in availability, load balancing may also be a requirement in building a highly available VPN.

As with all highly available solutions, a balance must be made between the level of availability required and the cost of implementing the solution. The cost could be monetary, but it could also take the form of slower performance, user inconvenience, implementation downtime, or in some other way. Each of these forms can also be expressed in terms of money, of course.