Subject | Re: [IBO] development |
---|---|
Author | Geoff Worboys |
Post date | 2010-03-23T02:04:14Z |
Hans wrote:
But anyway I can definitely see where there is a problem.
Using your precompiled executable I get: 30-ish seconds for
FIBPlus, 48-ish seconds for IBO. The biggest part of the
problem seems to centre around a couple tables that send
the CPU through the roof for quite a long period of time.
I will try with my optimisations in place, but while they
should help because there are so many almost empty tables,
I doubt if they will help with the particular high CPU
problems I see in the execution of your demo.
Chopping things down to isolate the problem will take some
time, I will get back to you.
--
Geoff Worboys
Telesis Computing
> I just sent you my little test source code plus database.OK, I've got the demo... would have liked something simpler.
But anyway I can definitely see where there is a problem.
Using your precompiled executable I get: 30-ish seconds for
FIBPlus, 48-ish seconds for IBO. The biggest part of the
problem seems to centre around a couple tables that send
the CPU through the roof for quite a long period of time.
I will try with my optimisations in place, but while they
should help because there are so many almost empty tables,
I doubt if they will help with the particular high CPU
problems I see in the execution of your demo.
Chopping things down to isolate the problem will take some
time, I will get back to you.
--
Geoff Worboys
Telesis Computing