Griggs, Donald wrote: > If you really have to have even more performance, you could cache > inserts in RAM and wrap multiple INSERTS with a single transaction when > you have a certain number cached (or if a max delay is reached). > shouldn't varying the transaction size have the same effect? moving from 4096 to 8192 didn't show any improvement.
_______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
