Hi Richard On Wed, Dec 5, 2012 at 9:05 AM, Richard Hipp <d...@sqlite.org> wrote: > The $dir will contain initial database contents and scripts of SQL > statements that were run. You can rerun those SQL statement using the > command-line shell to find slow ones, then do things like EXPLAIN QUERY > PLAN to figure out why they are slow and perhaps fix them by adding indices.
Thank you very much for your comprehensive reply. I actually know the queries, because the application logs both the SQL and times. They are simple SQL statements, eg "insert into table_a (...) values (...)" or "update table_a set aaa='...' where bbb='...'". I just tried enabling the write-cache of the drive, and the warn logs disappeared. So, I know that somehow the OS/filesystem/disk setup is blocking the inserts/updates quite severely for sqlite (but not for MySQL). Is there anything else I can do to improve the situation? Ideally I would like to understand what is happening under the hood of sqlite, to improve my understanding of the difference between the 2 DBs. Any other ideas for me to trace this at a lower level would be greatly appreciated. I think I will need to get deeper into linux's block layer stuff, so perhaps this is not the correct place to ask the question, but I thought I'd start here. Regards Keith _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
