[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [mat-dev] BHeapSampler memory analysis tool made public
|
Hi Arndt,
The Sampling approach always seemed interesting to me, but I only
thought about approximating certain memory usage indicators.
I do not 100% understand how you compute the the memory usage, could
you elaborate that a bit more?
Are you aware that MAT has a feature in the Dominator tree that allows
you to group by class (or package)?
I believe one could build an approximated "memory domination graph" by
using the data from this view.
Regards,
Markus
On Sun, Nov 13, 2011 at 2:00 PM, Arndt <Arndt.244488@xxxxxxxxxxx> wrote:
> Hello "mat-dev" list,
>
> first let me emphasize the great work you did on the Eclipse Memory Analizer
> Tool. I enjoy it.
>
> I recently made public a binary version of an inhouse development for
> analysis of large java heap dumps that I would like to present you. It's
> based on statistical sampling of memory-paths and the compilation of a
> class-level-graph showing the dominant structures.
>
> Im looking for a discussion of the conceptual differences between a
> class-level-graph and MATs dominator tree and it's leak report. I've some
> remaining issues in creating a reasonable display of memory held by
> different referer paths, which relates to your discussion on the shallow an
> retained memory, and I've an issue in calculating appropriate instance
> counts for graph-nodes with not class-identity. Maybe a constructive
> discussion can sort this out, and maybe it's a short hack to add a
> class-level-graph to MAT?
>
> Feedback welcome...
>
> Get my tool at: http://dr-brenschede.de/bheapsampler
>
> Regards,
>
> Arndt
>
>
> _______________________________________________
> mat-dev mailing list
> mat-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/mat-dev
>
