5.2. Clustering topologies and failover rules
There are many aspects to consider when designing and planning high availability with SQL Server clustering. Chief among them is the number of dedicated passive, or standby, servers to include in the cluster for failover purposes.
In simple clusters with a 1:1 working:standby ratio, each working server is matched with a standby server that, during normal operations, isn't used for any purpose other than to simply wait for a failure (or planned failover). Alternatively, both (or all) servers can be active, but this introduces resource complications during failovers; for example, will a server be able to handle its own load plus that of the failed server?
In previous versions of clustering, Active/Passive ...