I am using the internal "Profile" module to check the runtime complexity
of functions in unit tests.
"Profile.getInfo" returns a list of records like this:
profileStats(calls:1001 closures:1 column:0 file:'Oz' heap:25120 line:9
name:'ForProc' samples:0)
"samples" does not work, but I found that "calls" (number of function
calls) often reflects the used time quite accurately.
"heap" is useful to trace the memory consumption.
Example usage:
\switch +profile
declare
[Profile] = {Module.link ['x-oz://boot/Profile']}
Dict = {NewDictionary}
in
{Profile.mode true}
{Profile.reset}
for I in 1..1000 do
_ = I div 2
Dict.I := I
end
{Inspect {Profile.getInfo}}
Of course, this is an internal, undocumented feature, so use it at your
own risk ;-)
Cheers,
Wolfgang
Torsten Anders wrote:
Dear all,
I would like to compare the performance of a number of
(problem-specific) distribution strategies with a set of constraint
problems and with different settings (e.g., different recomputation
settings). I am interested in computation time (wall time) and memory
consumption.
Because I will end up with quite a lot of test cases to measure, I
would like to automate this process. Is there a recommended way for
doing such measurements?
Concerning runtime measurements, I plan to use the following function.
DO you see any problems with this strait-forward approach?
/** %% Returns the time (in msecs) the application of P (a null-ary
procedure) took.
%% */
fun {TimeSpend P}
Start End
in
Start = {Property.get 'time.total'}
{P}
End = {Property.get 'time.total'}
End - Start
end
My problem is that I don't know how I can test the memory consumption
automatically. I know about the Profiler, but as far as I can tell
from its documentation the Profiler is indented only for interactive
use, not for automatic measurements.
Any other ideas?
Thank you very much!
Best
Torsten
--
Torsten Anders
Interdisciplinary Centre for Computer Music Research
University of Plymouth
Office: +44-1752-233667
Private: +44-1752-558917
http://strasheela.sourceforge.net
http://www.torsten-anders.de
_________________________________________________________________________________
mozart-users mailing list
[email protected]
http://www.mozart-oz.org/mailman/listinfo/mozart-users
_________________________________________________________________________________
mozart-users mailing list
[email protected]
http://www.mozart-oz.org/mailman/listinfo/mozart-users
