NickPyner:
I submit you should just cut to what the real problem is, and forget about answering unanswerable questions, for which you probably already have a solution.
I submit that you need to sort out what data you're dealing with (i.e. the type and quantity) and what you want to do with it, before you design mechanisms to store it. These questions ought to be easy to answer, and if you can't answer them then what hope do you have of choosing the most appropriate design?