Thank you so much for the detailed explanation, Wrog. I do think the learning curve is steep if I'd like to implement a database for my particular research problem, given my current standing. however, i'd definitely make an effort to understand basics like how disks and memory work as i work along my projects, which i now feel is indispensable. the scales of my previous work just never blow up like this to remind me just how poorly my algorithms were written.
in reply to Re^5: how to merge many files of sorted hashes?
in thread how to merge many files of sorted hashes?
as for postgis, i have not taken a close look at it yet, and i'm not entirely sure that the database design permits the kind of flexibility i need for my actual application; i have many more attributes other than coordinates of the points that i need to take care of, so maybe coding it myself will actually be easier and faster (to meet my project deadline).
i just tried the method you suggested first and it went orders of magnitudes faster (as expected). i used this 100 point cloud to test the upper bound of the running time for all the point clouds, and with this largest one finishing within a day my calculations are now much more scalable. thank you so much!