Skip to Content.
Sympa Menu

cgal-discuss - [cgal-discuss] Re: Running out of memory when compiling 3D mesh generation code with extended polyhedral surface information

Subject: CGAL users discussion list

List archive

[cgal-discuss] Re: Running out of memory when compiling 3D mesh generation code with extended polyhedral surface information


Chronological Thread 
  • From: Nicholas Mario Wardhana <>
  • To:
  • Subject: [cgal-discuss] Re: Running out of memory when compiling 3D mesh generation code with extended polyhedral surface information
  • Date: Tue, 7 Jun 2011 12:31:34 +0800
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; b=DixZSde+39Ri9N4eAnU9NnQCTE3wgyBUOhqXmvZzyEbxQbDpqlmhXwCRfBS+SDHz9h XqZYgpPABx0g+fSuBdQnolqgax9Rr1Qks0PZfvU+YDvyiRm1fKLhrcIp17Isl8FH7osh TCpyacmBHi605IFZxhP0HfZm84UqaHJq5Vhbs=

On 7 June 2011 00:00, Nicholas Mario Wardhana
<>
wrote:
> Hi all,
>
> I am using CGAL's 3D mesh generation module (
> http://www.cgal.org/Manual/3.5/doc_html/cgal_manual/Mesh_3/Chapter_main
> ).to generate a 3D mesh from an extended polyhedral surface, and I was
> having a problem with the overwhelming memory consumption during the
> compilation. I tried to (independently) compile my code using Visual
> Studio 2008 on 2 machines:
>
> 1) laptop with Intel Core 2 Duo P8600 @ 2.40 GHz processor, 4 GB of
> memory (1 GB is shared for graphics), running Windows Vista, and
>
> 2) workstation with.Intel Xeon E5520 @ 2.27 GHz quad core processor, 4
> GB of memory, running Windows 7.
>
> Please see the attachment for a rather stripped off, albeit not
> minimal code. This code can compile well on both machines, although
> with high memory consumption. However, for the real code that we are
> developing, the second machine can only compile the code under Release
> configuration, whereas the compiler always crashes on the first
> machine. I also attach a screen capture of the Windows task manager
> during the compilation on the first machine. It shows that the
> compilation process consumed almost 2 GB of memory.
>
> My questions are as follow.
>
> 1) Is it normal for the compiler to consume that much memory? Is there
> any way to reduce the memory consumption?
>
> 2) What is the fastest kernel that the tetrahedral mesh generation can
> use? I only want a fast program without exact computation. Originally
> I went with CGAL::Simple_cartesian<float>, but I found an "cannot
> convert from 'const CGAL::Lazy_exact_nt<ET_>' to 'float'" error, so I
> resort to CGAL::Simple_cartesian<double> .
>
> Please let me know if I need to give more information.
>
> Thank you!
>
> Best regards,
> Nicholas Mario Wardhana
>

I'd like to make a few updates.

1) I use CGAL 3.8 (I gave an incorrect reference to CGAL 3.5 documentation)

2) I tested the attached code again just now, and the second machine
can only compile under Release configuration. Please see attachment
for a capture of Windows Task Manager.

3) I tried to comment the lines in the main function one by one, and
apparently the culprit is the following line:

MeshComplex3 meshComplex3 = CGAL::make_mesh_3<MeshComplex3>(domain, criteria);

Meanwhile, I am currently re-reading the documentation again, in case
there is something that I have missed. Thank you.

Best regards,
Nicholas Mario Wardhana

Attachment: cgal2..jpg
Description: JPEG image




Archive powered by MHonArc 2.6.16.

Top of Page