ActiveMQ is not ideally suited for our use case, this ticket is for spiking out the performance, usability and simplification that we could gain by switching to more of a qmail like, file per command model. This ticket is entirely focused on the storage of commands and not concerned (yet) with how those will be processed. What is covered in this is adding/creating/spiking how command metadata would be stored and the performance impacts of that information on startup time etc.