Hello sqlite users !
I have noticed that some people (like me) want to use sqlite in a read only
environment.
Basically, we only do select commands, but we want them to be as fast as
possible.
A bit like CDB, but with SQL queries.
Is there any compilation hint or any way to hack sqlite to make it faster in
such an environment (get rid off the transactions for example) ?
For the moment, I
- have synchronuous off,
- re-use queries prepared in advance.
- have only one big transaction for the whole run
Also,I was surprised to see that sometimes the cache is not efficient.
For example, if I do:
SELECT A.x, B.y
FROM A,B
WHERE A.col1='VALUE'
AND A.col2 = B.col3;
With indexes
Idx1: A.col1 , A.x
Idx2: B.col3, B.y
and execute it several times, I get almost always the same response time
(around 1.5 ms in my case).
Is sqlite only relying on the disk cache for the data or am I missing something
here ?
Cheers !
Renaud
_________________________________________________________________
Discover the new Windows Vista
http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE
