The generator produces jobs based on a configured interarrival
distribution. A normalized thinning function is used to enable
non-stationary job generation. For non-stochastic generation of jobs, a
random variable distribution with a single point can be used - in which
case, the time between job generation is constant. This model will
produce jobs through perpetuity, and the generator does not receive
messages or otherwise change behavior throughout a simulation (except
through the thinning function).