Hello everyone, I tried to get some answers using the mailing list archives, but you guys have been communicating so much on this, it's hard to run through everything that was written ;)
Basically, I'm using sqlite3 in my C application. I believe the tool SQLiteBrowser is using it as well. I noticed that the difference of query time execution between my app and the browser, with the same queries of course, can be widely different. In fact, it feels like it's exponential. The difference is really tiny for fast queries, but i have some queries that take 10 seconds of execution in SQLiteBrowser, and 100 seconds in my application. Context : the databases i'm working on aren't large at all (between 10Mb and 30 Mb, no more than 50 Mb for sure). Tables im working on don't exceed 100k rows, which is kind of small when i compare to some of the things i read on the mailing list. In order to measure that execution on time on my app, I've surrounded the exec() function of sqlite3 with a clock() mechanism. To finish, I'm working on a in-memory database, with those pragmas : page_size = 4096, cache_size = 10000, thread safe, automatic index = on, synchronous = off, journal_mode = MEMORY", locking_mode = exclusive, temp_store=memory. The reason why for those pragmas is that I'm essentially running SELECT queries, a very few INSERT or UPDATE but I'm mainly reading in the database. I've tried to check the compile options of SQLiteBrowser to begin with, but I was unable to find them. Anyway, I was wondering why is there such a difference, I feel like the pragmas are ok, and I've tested with same database, same queries. Any ideas? Thanks for reading, if you guys need to know more I'm here -- Laura BERGOENS Technicienne supérieure en Informatique et étudiante à l'IMERIR de Perpignan *Institut Méditerranéen d'Étude etde Recherche en Informatique* _______________________________________________ sqlite-users mailing list sqlite-users@mailinglists.sqlite.org http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
