[gmx-developers] Alternate Parallelization scheme

[David]

On Sat, 2005-06-18 at 15:41 -0500, Nathan Moore wrote:
> >> (2) Which particles does each node control?  I assume that the grompp
> >> options -sort and -shuffle destroy the easy mapping, rank=0 gets the
> >> first
> >> 10 atoms, rank=1 gets the second 10 atoms etc.
> > This doesn't matter once you're in mdrun. All this stuff is done in the
> > preprocessor. As a matter of fact mdrun prints in md0.log how the
> > division over processors is done.
> 
> How is the division of labor coded into mdrun?  Is there an array that
> maps the control of atoms to each processor?  I'm imagining an array,
>      int control[num_atoms];
> where the value of control[i] is the rank of the processor assigned to
> update force and position of atom i.
t_nsb
(Neighbor Search Border)

We have a project going to include dynamics load balancing in the code,
but it is not completely trivial, and the main developers (Erik, Berk,
me) have not enough time...
-- 
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://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++