On Thu, 9 Dec 2010, Michael Povolotskyi wrote:

> The idea to use files sounds interesting, but I'd like to try  the 
> distributed creation first.
> If it works, I'll write a report to the maillist and, may be, someone else 
> founds it useful.

Definitely.  In fact, before writing a mesh generation function
specific to your app, you may want to start by writing a parallel
version of MeshGeneration::build_cube()... your incentive would be
making it easier to "practice" the problem and catch any bugs, but my
ulterior motive for asking is that I'd love to get a patch adding such
capability to the library as an example for others.

> 1) In the ParallelMesh class the functions reserve_nodes or reserve_elem do 
> nothing, right?

Right.

>    Is there any  way to preallocate memory for points and elements?

No, the underlying container is a vector-like interface around
std::map (which has no preallocation capability).  It might be worth
eventually writing a custom container for better performance; I don't
think even ropes have a standard STL implementation.

> 2) Is the following statement correct:
> If, after adding all the nodes and all the elements, I set the partitioner 
> pointer to NULL and call the prepare_for_use() function,
> then the mesh will be ready for use.

Yes, but only if you've already manually set processor IDs on all
elements and consistent proc IDs on all nodes.
---
Roy

------------------------------------------------------------------------------
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users

Reply via email to