[gmx-users] domain decomposition and load balancing

[Mark Abraham]

----- Original Message -----
From: Amit Choubey <kgp.amit at gmail.com>
Date: Saturday, February 20, 2010 10:25
Subject: Re: [gmx-users] domain decomposition and load balancing
To: Discussion list for GROMACS users <gmx-users at gromacs.org>

> Hi Mark and Justin,
> I am using a box of size (20nm)x(20nm)x(60nm)..system is open along z ie (60nm) side.> 
> I tried using only two nodes, it gives the same error> 

> There is no domain decomposition for 2 nodes that is compatible with the given box and a minimum cell size of 0.889862 nm> Change the number of nodes or mdrun option -rdd or -dds> Look in the log file for details on the domain decomposition

... so what does the log file say?

Mark

> On Fri, Feb 19, 2010 at 3:16 PM, Mark Abraham <mark.abraham at anu.edu.au> wrote:

> ----- Original Message -----
> 
From: "Justin A. Lemkul" <jalemkul at vt.edu>
> 
Date: Saturday, February 20, 2010 10:13
> 
Subject: Re: [gmx-users] domain decomposition and load balancing
> 
To: Discussion list for GROMACS users <gmx-users at gromacs.org>

> 
> 
>
> 
>

> > Amit Choubey wrote:
> 
> >Hi Mark,
> 
> >
> 
> >I am not using PME calculation.
> 
> >
> 
> >I was hoping mdrun will do the cell allocation itself.
> 
> >
> 
>
> 
> It will, unless it can't, which is exactly your problem. 
> 
> Mark's point stands, regardless of whether or not you're using
> 
> PME.  DD requires certain minimum size requirements (which
> 
> are discussed in the manual and the Gromacs 4 paper), so you
> 
> have two choices:
> 
>
> 
> 1. Read about the options mdrun is telling you about.
> 
> 2. Use fewer nodes so that the DD algorithm can construct
> 
> reasonably-sized domains.

> 

> 3. Use a larger simulation system so the minimum cell size is larger w.r.t. the cutoffs
> 
4. (If reasonable) Don't use pbc=xy

> 
> 
Mark
> 
--
> 
gmx-users mailing list    gmx-users at gromacs.org

> http://lists.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 gmx-users-request at gromacs.org.
> 
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

> 

> -- 
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.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 gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php