Thanks, I am trying to install Trilinos these days in Ubuntu-12, but
still has not been successful yet. The dependences and paths seem
complicated.
I have a new question about the noise term. I would like to have
Langevin noise terms(please see attached picture"Langevin noise
term.png"). I noticed it is very similar to Fipy GaussianNoiseVariable
as shown in the Fipy web
http://www.ctcms.nist.gov/fipy/fipy/generated/fipy.variables.html
My questions are:
(i) Since I understand little about statistical mechanics, I am not sure
if Langevin noise term in the picture is also "uncorrelated in space and
time" as describe in Fipy web?
(ii) How can I implement the noise in equation (1a) in the picture?
There is a Laplacian operator, so I guess
sigmaSqrd must be something different.
On 03/03/15 17:31, Daniel Wheeler wrote:
On Mon, Mar 2, 2015 at 4:21 PM, <[email protected]> wrote:
Dear Fipy Developers,
When I try to test the parallel running by $ mpirun -np 3 python
examples/parallel.py. I got the following error massage.
from PyTrilinos import Epetra
ImportError: No module named PyTrilinos
If I understand correctly, Trilinos is a huge package with many
sub-packages(including PyTrilions and Epetra). I do not want to install
all of the Trilinos. So, I am wondering if there is the list of minimum
required packages for parallel running regarding the Trilinos?
I think Jon listed out what we normally specify for cmake when
installing Trilinos to work with FiPy. I don't worry about the
quantity of packages installed. I think the duration of the
installation is much more concerning and I don't believe that it is
that long for Trilinos anymore. Last time I installed Trilinos it
didn't feel that long probably less than an hour. FiPy only uses two
packages in Trilinos, I believe, but it really isn't worth setting up
the cmake script to optimize this. Something will just go wrong and
you'll spend hours debugging it. There are all sorts of dependencies
within Trilinos so there isn't much point optimizing unless you really
know Trilinos well IMO.
--
------------------------------------------
Ronghai Wu
Institute of Materials Simulation (WW8)
Department of Materials Science and Engineering
University of Erlangen-Nürnberg
Dr.-Mack-Str. 77, 90762 Fürth, Germany
Tel. +49 (0)911 65078-65064
_______________________________________________
fipy mailing list
[email protected]
http://www.ctcms.nist.gov/fipy
[ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
