http://www.ca.postgresql.org/docs/faq-english.html#4.8
If you add "VACUUM ANALYSE" at the end of your script, your script will
always take the same amount of time, but it will likely be a tad longer
than two seconds.
At 04:52 PM 1/29/02 -0800, Cheirie Shum wrote:
>Hi, I've written a little tcl program to test how many seconds it takes to
>run an 'update' transaction and found that the running time increases by a
>couple of seconds every time i run my script.
>
>My script does the following:
> -connect to postgres
> -updates a row and then updates it again with some other value
> -iterate the update step above 200 times
> -print the number of seconds that iteration took
> -disconnect from postgres
>
>Results: So the first time, i get about 2 seconds, then 4, 6, 8... etc.
>
>So, each time that row gets updated, the longer the next update takes...
>What is happening? What do i need to do so that postgres doesn't do this?
>... What do i need to do so that postgres always updates a row 200 times
>in 2 seconds?
>
>Any help would be appreciated.
>Please reply to: cshum@cats.ucsc.edu
>
>
>
>
>---------------------------(end of broadcast)---------------------------
>TIP 5: Have you checked our extensive FAQ?
>
>http://www.postgresql.org/users-lounge/docs/faq.html
>
