On Tuesday, 17 December 2013 at 19:09:49 UTC, H. S. Teoh wrote:
What's a good, efficient file structure for storing extremely large lookup tables? (Extremely large as in > 10 million entries, with keys and values roughly about 100 bytes each.) The structure must support efficient adding and lookup of entries, as these two operations will bevery frequent.
But 200*10million = 2GB. Can't you use an existing KD-tree and tweak it to fit memory pages and rely on paging for a start?
