You'd have the same amount of data either way, right? The memory for the containers themselves would be relatively small.
I guess it depends on how you define "easier". I personally keep a master list of data, and then have a sorting routine that breaks it down into smaller parts. My table view methods only care about the smaller parts; they don't care how the data was sorted.
Here is something that should help:
Clinging To Ideas: Taming Table Views
After I wrote that, I learned that CoreData has a NSFetchedResultsSectionInfo class which is basically the same thing, so I think I'm on the right track.
And you only "need" 2 arrays: master and sorted.
