Hi there

Basically on one server in the cluster can own and work with a set of
resources at a point in time. This doesn't mean that node 1 and node 2
cannot be active though, they can be, just with different clustered
resources; the resources will swap between nodes as need be based on XYZ
failover condition of course. A classic example in the sql server space, is
that node 1 hosts my OLTP db, and node 2 hosts my reporting server, if node
1 fails then node 2 takes up the OLTP and the Reporting db, of course
hammering performance if you are not too careful

Cheers
CK

MS Sql Server LazyDBA home page