> -----Original Message-----
> From: Tim Anderson [mailto:[EMAIL PROTECTED] 
> Sent: 10 January 2004 11:40
> To: [EMAIL PROTECTED]
> Subject: [sqlite] .Net provider performance

> The performance of the three .Net providers is 
> near-identical. On the first run, with the system restarted 
> to eliminate any caching by the file system, it takes about 4 
> seconds to execute the query and fill a datagrid with the 
> results. On subsequent runs it takes about 0.25 seconds, 
> showing the dramatic effect of the caching.
>  
> The JET provider is an interesting case. The first run takes 
> about 9 seconds. However, subsequent runs are about 0.15 seconds.

This result bothered me so I tried to improve it. I added a new compound
index and got a big improvement from Sqlite, less so from Jet:

1st run Sqlite 1.2 secs
Subsequent run Sqlite 0.02 secs
No real difference between the various providers.

1st run JET 7.8 secs
Subsequent run JET 0.13 secs

So now either way Sqlite is around 6 times faster *in this particular
case* :-)

Tim

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to