Michael: While my answer isn't strictly limited to sqlite, the performance nut in me can't help myself. You have a a table with only one column, a string. I could be wrong here but it seems like you just want to keep a list of values that you have already tried. After you insert a bunch of strings into the table you want to be able to quickly look up a string to see if it exists, so that you can tell if you already probed that sequence (taking a guess here).
If my guess is right, and the only thing you are doing is looking up sorted single-column elements, you probably can get away with your own quick disk-based binary tree implementation and avoid sqlite for this particular circumstance altogether. The result would be several orders of magnitude faster, even after you have followed the suggestions others have given. > Hi, > > I am new with sqlite, and I create a program that reads > several mllion > records and puts them into a sqlite db using. > > The table has one column ONLY indexed and unique, but it > takes many hours. > > Is it necessary to pre-allocate the space, or is the > anything that I can > do to reduce the time it takes. > > this is how I create the db, table and index. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
