>Spark is a terribly inefficient solution to any known/stable data processing or analytics jobs.

Can you expand on that?