Your organization deploys a client$-$facing application to a series of virtual machines fronted by a load balancer. The application is stateless, so any request may be routed to any server. You use auto$-$scaling to ensure that there are always enough servers to meet demand, and you specify that there are never to be fewer than two servers provisioned $($even when demand is very low$)$ because the application must remain available even if one VM goes down. Which load$-$balancing strategy should you employ?

Equitable dispatching

Hash$-$based distribution

Capacity$-$based distribution

None of these strategies increases an application's availability. The best strategy for increasing availability is increasing the number of servers.