Jens Pohl wrote:
Hello!
I have a similar problem with installing gromacs on a dual core machine.
When I go through the procedure described by Matheus Fazian Ige Gondo a bit further down installing lam/mpi then fftw and finally gromacs everything seems to be working fine, but lam/mpi doesn't recognize the two cores and with mdrun there will be an fatal error. grompp and mdrun both were run with -np 2 and the output of mdrun says, that the file was made for two nodes but mdrun expected it to be for one node.
With lamnodes there is only one node on cpu0 on localhost... What to do in this
case?
Fetch fftw 3.1.1 (latest stable)
./configure --enable-float
make install
Fetch openmpi 1.1.x (latest stable)
./configure
make install
Fetch gromacs
./configure --enable-mpi
make install
That's all you need
Thanks in advance
Jens
Hello Ricardo
Did you install with the sources ? (make install) ?
Go to the folder where you ran this command and run "make uninstall"
Matheus Fazian Ige Gondo
Departamento de Física e Biofísica
Instituto de Biociências
Unesp - Botucatu - SP
[EMAIL PROTECTED]
+55 (14) 3811-6254
On Tue, Mar 11, 2008 at 1:55 PM, Ricardo Soares <[EMAIL PROTECTED]> wrote:
Kpiwara De X-nelo wrote:
Hello
You'll need FFTW (fftw.org) LAM/MPI (lam-mpi.org) and gromacs sources
Download the 3 packages
start with the Lam/MPI (configure , make , make install)
Then fftw (the 2.1.5 version is needed, the 3.2 supports it but is
still unstable) (configure --enable-mpi, make, make install)
Then, finally, the GROMACS (configure --with-fft=fftw2 --enable-mpi ,
make , make install)
It should be running
as a non root user type lamboot
It should start the lam/mpi
run your simulations as usual just changing these
grompp -v -f xx -o xx -c xx -p xx -np
#
the -np generate statusfile for
# nodes
then
mpirun c0 C
mdrun
-v -s xx -o xx -e xx -c xx -g x -np
# log >& xx
Number of nodes (#) , must be the same as used for
grompp
the mpirun c0 C tells it to run on the maximun available nodes.
Hope I helped.
Matheus Fazian Ige Gondo
Departamento de Física e Biofísica
Instituto de Biociências
Unesp - Botucatu - SP
[EMAIL PROTECTED]
+55 (14) 3811-6254
Hello Matheus, you sure did help!
But that way I'd need to uninstall gromacs and reinstall. Is there a
way to include the LAM/MPI package in an already installed instance of
Gromacs? If not, how do I uninstal it?
Thanks again!
-- ___________________________________________________________
Ricardo Oliveira dos Santos Soares
Post-graduation Student in Biological Physics
University of Sao Paulo - USP
Faculty of Farmaceutical Sciences of Ribeirao Preto - FCFRP
Phone: 55 (16) 3602-4840
Curriculum Lattes - http://lattes.cnpq.br/0777038258459931
___________________________________________________________
_________________________________________________________________________
In 5 Schritten zur eigenen Homepage. Jetzt Domain sichern und gestalten!
Nur 3,99 EUR/Monat! http://www.maildomain.web.de/?mc=021114
_______________________________________________
gmx-users mailing list gmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php
--
David van der Spoel, Ph.D.
Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone: +46184714205. Fax: +4618511755.
[EMAIL PROTECTED] [EMAIL PROTECTED] http://folding.bmc.uu.se
_______________________________________________
gmx-users mailing list gmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php
