Hi,
On 05/12/2009 12:46 AM, Dimitri wrote:
> So, why I don't use prepare here: let's say I'm testing the worst
> stress case :-) Imagine you have thousands of such kind of queries -
> you cannot prepare all of them! :-) or you'll maybe prepare it once,
> but as I showed previously in this thread prepare statement itself
> takes 16ms, so for a single shot there is no gain! :-)
I have a hard time imaging a high throughput OLTP workload with that
many different queries ;-)
Naturally it would still be nice to be good in this not optimal workload...
Andres
