On Oct 21, 2010, at 1:00, Kees Nuyt wrote:
> PRAGMA page_count; and PRAGMA freelist_count; will give you
> some info, but not as much as sqlite3_analyzer.
> It might be enough in your case.
Those provide some info, but not the specific info I'm having problems
with right now. I have too many non-sequential pages and it's making my
application run a couple of orders of magnitude slower than a fresh DB.
> Running a full analysis takes quite some time, you might as
> well start a VACUUM periodically. I would suggest to run it
> with half the frequency you intend to run the analysis. That
> way the time spent to this kind of housekeeping will be
> about the same.
We're not measuring it taking too long at this point. We have a lot of
application-specific optimizations we could perform if we had a bit more
information.
> And you could consider
> PRAGMA auto_vacuum = 0 | NONE | 1 | FULL | 2 | INCREMENTAL;
The problem with auto_vacuum is that it's documented to make the
problem worse.
--
Dustin Sallings
_______________________________________________
sqlite-users mailing list
[email protected]
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
