[gmx-users] Gromacs on Pathscale/Infiniband?

[Florian Haberl]

Hi,

On Wednesday 12 July 2006 21:26, Susan Chacko wrote:
> Hi all,
>
> We have a bunch of Opterons connected with Infiniband. I'm trying to
> get Gromacs compiled for these nodes. I successfully built FFTW
> 3.1.2, but my Gromacs 3.3.1 config goes about halfway and then fails
> with
>
> checking size of int... configure: error: cannot compute sizeof
> (int), 77
>
> These are the flags I'm using:
> ./configure --with-fft=fftw3 --enable-float --prefix=/usr/local/src/
> gromacs/gromacs-ib --enable-mpi
>
> Any suggestions? Has anyone else successfully built/run Gromacs with
> Pathscale/IB?

We are running here gromacs on infiniband and xeon architecture without 
problems,  i used gcc and intel compiler, the performance is nearly the same 
(0,5% more or less).

Disable fortran for compiling, we are using mvapich (you can get it here: 
http://nowlab.cse.ohio-state.edu/projects/mpi-iba/ ) for mpich communication.

>
> All help much appreciated,
> Susan.
>
> ------------------------------------------------------------------------
> ----------
> Susan Chacko
> Computational Biologist, Helix Systems
> Center for Information Technology
> 12B/2N207                                               Ph: 301-435-2982
> National Institutes of Health                           Fax:
> 301-402-2190
> Bethesda, MD 20814                                      Email:
> susanc at nih.gov
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php


Greetings,

Florian

-- 
-------------------------------------------------------------------------------
 Florian Haberl                        
 Computer-Chemie-Centrum   
 Universitaet Erlangen/ Nuernberg
 Naegelsbachstr 25
 D-91052 Erlangen
 Mailto: florian.haberl AT chemie.uni-erlangen.de
-------------------------------------------------------------------------------