This is a continuation of the "SQLite vs. Oracle (parallelized)" thread with a request to learn how others are using SQLite with very large data sets. The context of this post is processing large data sets from a single process perspective, eg. this question is being asked from a batch data processing vs. multi-user perspective. 1. In browsing the archives, it seems that one technique is to split or partition large data sets into separate SQLite databases that can be loaded and indexed independently of one another (possibly via separate processes on the same box or on separate boxes). It appears that some people have written their own front-ends to manage how records are inserted and/or read from a collection of SQLite databases. 2. Another technique appears to be to run SQLite on boxes with lots of memory and then configure SQLite to make optimal use of available memory. Are there other techniques that one should consider and/or what techniques should one avoid? Thank you, Malcolm _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users