[gmx-users] Domain Decomposition

Iman Ahmadabadi imanahmadabadi75 at gmail.com
Thu Feb 15 17:39:22 CET 2018


Dear Gromacs Users,

In one job, I always get in (any number of nodes) the domain decomposition
error as following and I don't know what should I do. I have to use the
-dds or -rdd setting for my problem?

Sincerely
Iman

Initializing Domain Decomposition on 56 nodes
Dynamic load balancing: auto
Will sort the charge groups at every domain (re)decomposition
Initial maximum inter charge-group distances:
    two-body bonded interactions: 9.579 nm, LJ-14, atoms 1663 1728
  multi-body bonded interactions: 9.579 nm, Angle, atoms 1727 1728
Minimum cell size due to bonded interactions: 10.537 nm
Maximum distance for 5 constraints, at 120 deg. angles, all-trans: 0.700 nm
Estimated maximum distance required for P-LINCS: 0.700 nm
Guess for relative PME load: 0.87
Using 0 separate PME nodes, as guessed by mdrun
Scaling the initial minimum size with 1/0.8 (option -dds) = 1.25
Optimizing the DD grid for 56 cells with a minimum initial size of 13.171 nm
The maximum allowed number of cells is: X 0 Y 0 Z 0

-------------------------------------------------------
Program mdrun, VERSION 4.6
Source code file: /share/apps/gromacs/gromacs-4.6/src/mdlib/domdec.c, line:
6767

Fatal error:
There is no domain decomposition for 56 nodes that is compatible with the
given box and a minimum cell size of 13.1712 nm
Change the number of nodes or mdrun option -rdd or -dds
Look in the log file for details on the domain decomposition
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------


More information about the gromacs.org_gmx-users mailing list