That SNFAIL example had four overlapping events analyzed together, each with their own load vector and timing. You can include several events. There is no formal limit on the overall duration of the analysis but you may run into memory problems if you try to run a very long or a very detailed (short time step) analysis.

We are seeing more users collecting time history data from transient hydraulic analyis packages and pushing those results through CAESAR II. My suggestion is to start simple and build from there. Don't try to hit the home run on the first pass.