Two things: 1. The longer the table names, the longer it will take to compute the hash of each table name.
2. Because the entire schema must be reprocessed after each change, all the table names will be rehashed after each table has been created. Creating 10,000 tables will result in re-reading all that data and re-hashing all the table names. After adding the 10,000th table, SQLite will have computed at least 50,005,000 hash operations. Many more if column names are hashed too. SDR On Sep 6, 2013 2:00 PM, "Harmen de Jong - CoachR Group B.V." < har...@coachr.com> wrote: > On 6 sep. 2013, at 20:09, "Kevin Benson" <kevin.m.ben...@gmail.com> wrote: > > Dr. Hipp does a little bit of explaining on this topic, generally, in his > > replies on this thread: > > > > http://www.mail-archive.com/sqlite-users@sqlite.org/msg78602.html > > Thanks for pointing me to that thread, but as dr. Hipp states in this > thread, the tables are stored in a hash. Therefore I would not expect a > large performance decrease on large number of tables at all, or am I > missing something? > > Best regards, > Harmen > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
