Probably the simplest way to do this is to do a select using your "new" data minus any information that might have changed (like the price). Then, if you get a result back, do an update. If not, then do an insert. Pretty simple, I think.

As for the speed issue, I'm fairly sure that this is an indexing issue. I routinely do inner joins of tables with 60 million rows in mysql on a web-based app. For best performance, any column that appears in a where clause or a join clause should be indexed, in general.