A database may not be the best solution here -- from reading the other posts, it could be that you're going to be more interested in 'clumping' each of the data points together, creating 'neighborhoods' of 'nearest neighbors'. My Systems Design professor Ed Jernigan did research along those lines.
Perhaps a first cut would be some sort of encoding of each data point, then a 'clumping' based on that, with further analysis on the smaller 'clumps'.
Alex / talexb / Toronto
"Groklaw is the open-source mentality applied to legal research" ~ Linus Torvalds
