[Pharo-project] Howto measure memory footprints of functions/objects/libraries in Pharo?

Mariano Martinez Peck marianopeck at gmail.com
Tue Nov 9 16:52:39 CET 2010


Maybe doing a SpaceTally new printSpaceAnalysis
and a Smalltalk garbageCollect.

before each test....and then compare the different files..

cheers

mariano

On Tue, Nov 9, 2010 at 12:46 PM, Max Leske <maxleske at gmail.com> wrote:

> Um... I should be able to get a visualisation by running somthing like this
> right (I only installed it, didn't subclass anything)?
>
> MemoryProfiler
>        viewProfiling: [ FSDiskFilesystem current working / foo ].
>
> I get a window but there's nothing in it...
> Also, is selecting the expression in the block above, right clicking and
> selecting "memory profile it" equivalent to sending #viewProfiling: ?
>
> Reading your paper I get the idea that this should work, when looking at
> the instructions on the moose site it rather seems that I need to subclass
> MemoryProfiler to actually profile anything.
>
> Cheers,
> Max
>
>
>
> On 09.11.2010, at 15:02, Alexandre Bergel wrote:
>
> > Ok, but please, provide feedback.
> >
> > Alexandre
> >
> >
> > On 9 Nov 2010, at 10:40, Max Leske wrote:
> >
> >> Sounds like what I just needed! I'll try it out.
> >>
> >> Cheers,
> >> Max
> >>
> >>
> >> On 09.11.2010, at 12:42, Alexandre Bergel wrote:
> >>
> >>> Hi Guido,
> >>>
> >>> We wrote a profiler to keep track of memory consumption.
> >>>
> >>> http://bergel.eu/download/papers/Berg10g-MemoryProfiling.pdf
> >>>
> >>> It will be presented at Smalltalks next week.
> >>>
> >>> Cheers,
> >>> Alexandre
> >>>
> >>> On 8 Nov 2010, at 23:47, Guido Stepken wrote:
> >>>
> >>>> E.g.: Three different solutions for a Euler problem:
> >>>>
> >>>> (1 to: 999) inject: 0 into: [:sum :i | (i \\ 3 = 0) | (i \\ 5 = 0)
> >>>>            ifTrue: [sum + i]
> >>>>            ifFalse: [sum]]
> >>>>
> >>>>
> >>>> (1 to: 999) detectSum: [:i | (i \\ 3 = 0) | (i \\ 5 = 0) ifTrue: [i]
> ifFalse: [0]]
> >>>>
> >>>> (1 to: 999) detectSum: [:i | i * (#(0 0 1 0 1 1 0 0 1 1 0 1 0 0 1)
> atWrap: i)]
> >>>>
> >>>> How do i measure the exact memory footprint of each solution?
> >>>>
> >>>> Pharo claims having reduced memory footprint at many places. Can that
> be proofed, e.g. with one comparable demo app, once run in old squeak, once
> in Pharo 1.1/1.2?
> >>>>
> >>>> And no, MessageTally spyOn: does not do right!
> >>>>
> >>>> Tnx in advance, Guido Stepken
> >>>>
> >>>>
> >>>
> >>> --
> >>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> >>> Alexandre Bergel  http://www.bergel.eu
> >>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>
> >>
> >
> > --
> > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> > Alexandre Bergel  http://www.bergel.eu
> > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
> >
> >
> >
> >
> >
> >
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gforge.inria.fr/pipermail/pharo-project/attachments/20101109/963d245d/attachment.htm>


More information about the Pharo-project mailing list