Phani,
On Monday, April 2, 2018 at 4:59:02 PM UTC-4, Phani Motamarri wrote:
>
> I am trying to modify the function "SparseILU::initialize" inside
> sparse_ilu.templates.h to call the mkl function "dcsrilu0" which requires
> the use of some of the data like ia, ja accessed inside "initialize"
On 04/02/2018 02:59 PM, Phani Motamarri wrote:
I am trying to modify the function "SparseILU::initialize"
inside sparse_ilu.templates.h to call the mkl function "dcsrilu0" which
requires the use of some of the data like ia, ja accessed inside
"initialize" function. The tricky part is that
Dear Prof. Wolfgang,
I am trying to modify the function "SparseILU::initialize" inside
sparse_ilu.templates.h to call the mkl function "dcsrilu0" which requires
the use of some of the data like ia, ja accessed inside "initialize"
function. The tricky part is that the sparsity pattern stores the
Yes, that is completely true. Information exchange is one of the main
aspects that plagues the domain decomposition methods. But probably this is
improved by the optimized Schwarz which introduces artificial boundary
conditions to enhance the information exchange. I am not sure how this
scales
In the step-3 example
(https://www.dealii.org/8.4.1/doxygen/deal.II/step_3.html), I noticed
that at first the variable phi is used as a test function. It is in the
same space with u, except for that the value is 0 on the boundaries. (in
this problem the boundary values for u is also 0, but
Jie,
I am solving time-dependent Navier-Stokes equations with varying
Dirichlet boundary conditions. Although I have read the "Constraints on
degrees of freedom" for many times, I am still not sure what the best
way to implement changing boundary conditions is. My time scheme is
fully
Thank you for your reply. I understand that DD methods are not quite
that popular now. But I wanted to explore the avenue keeping in mind the
issue of scalability as well. For scalability, as you mention
communication is probably the bottleneck.
No, communication is not the issue. It's not
Dear Wolfgang,
Thank you for your reply. I understand that DD methods are not quite that
popular now. But I wanted to explore the avenue keeping in mind the issue
of scalability as well. For scalability, as you mention communication is
probably the bottleneck. A few papers have worked towards
Hello everyone!
This is deal.II newsletter #26.
It automatically reports recently merged features and discussions about the
deal.II finite element library.
## Below you find a list of recently proposed or merged features:
#6144: Bug fix for graph coloring/graph partitioning using Zoltan
Pratik,
deal.II has the parallel::distributed::Triangulation class which
provides the functionality to partition and distribute the meshes using
p4est's partition and adaptive space filling techniques. According to
what I have understood from step-40, using the
On 04/02/2018 11:48 AM, Phani Motamarri wrote:
We have an ongoing research project with Caltech where we are trying to
explore ways of approximating the subspace spanned by k lowest eigen
values of a given positive definite matrix(FEM discretized matrix) in a
sparse format using the incomplete
We have an ongoing research project with Caltech where we are trying to
explore ways of approximating the subspace spanned by k lowest eigen values
of a given positive definite matrix(FEM discretized matrix) in a sparse
format using the incomplete Cholesky factor with reduced computational
Hello all,
I wanted to know if it was possible to do overlapping decompositions using
deal.II and p4est. These are my thoughts:
deal.II has the parallel::distributed::Triangulation class which provides
the functionality to partition and distribute the meshes using p4est's
partition and
13 matches
Mail list logo
