[gecode-users] Analysing search performance

[Lars Otten]

On 03/02/06 14:44, Christian Schulte wrote:
> You can build Gecode also as static libraries: then profiling should be
> easier.

I had so far only tried 'hacking' the Makefile by hand, by adding -static as
a compiler flag -- which made the compilation fail at some point.

I have, however, just remembered the flags to ./configure (--set-static=yes
--set-profile=yes), but although that finishes compiling the profile still
lacks the respective elements? I might try some more tomorrow...

> And please make sure that the search space is really the same! Count the
> nodes in both approaches and check that they are the same or at least in the
> same order of magnitude. Otherwise you might be hunting ghosts.

I will try to meet with Mattias soon and discuss that with him..

> Another thing is recomputation: do you use it? Is memory an isssue for your
> application?

The biggest instance I'm currently testing is eating about 1 GB of memory
when run, so on my machine it is an issue, yes. ;) I am not using
recomputation at the moment (for direct comparison with my randomized search
algorithm, which don't use recomputation either), but as I found that mostly
increases memory consumption anyway...

Just tonight, however, I have made some progress by toying with the ICL for
the distinct propagators that are posted. Especially in cooperation with
some of the random stuff I implemented, a solution seems to be found rather
quickly for most instances -- I'm quite happy about that at the moment! :))

Time to catch some sleep, thanks for your support so far,

/Lars