On 17 Feb 2009, at 15:34, [email protected] wrote: > Is there any advantage to having data pre-sorted (in index order) > before loading it into SQLite and indexing it? Or does indexing > work best if the index values are randomly ordered at data load > time?
From my own experience, there's a big performance gain both at the indexing stage and in queries that have to scan relatively large parts of the database in index order (at least if you're working with a database that doesn't fit in RAM). Of course, if you're indexing multiple columns, you can only pre-sort data for one of them, so on average there will be little gain. Best regards, Stefan Evert [ [email protected] | http://purl.org/stefan.evert ] _______________________________________________ sqlite-users mailing list [email protected] http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
