[gmx-users] CVS version

[David van der Spoel]

Sukit Leekumjorn wrote:
> Dear Gmx-developer
> 
> This is regard to the CVS version.  I came across Erik Lindahl's 
> presentation on Simulations of Membranes and Membrane Proteins from 
> Gromacs Workshop at CSC.  At the end of the presentation, he mentioned 
> that the membrane simulations can be scale up to >1000 CPUs in CVS.  If 
> anyone have try this on the CVS version, could u please give me 
> information on how to run this version of Gromacs.  I see that there are 
> new options implemented in "mdrun" and -shuffle is removed from 
> "grompp".  Below are the new options in "mdrun".
> 
> -dd vector  1 1 1  Domain decomposition grid, 1 is particle dec.
> -npme    int      0  Number of separate nodes to be used for PME
> -[no]cart   bool     no  Use a Cartesian communicator
> -rdd   real      0  The minimum distance for DD communication
> -[no]dlb   bool     no  Use dynamic load balancing
> 
> 
just try e.g. mdrun -d 8 8 8 on 512 processors and play with the other 
flags. Since these flags are by no means the final ones (that is, we 
would like to automate things further) it doesn't make sense to get into 
a lot of detail. Also, please beware that this is  a development version 
and check you results.
-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++