[gmx-users] Restart simulation from checkpoint file with fewer nodes

Justin Lemkul jalemkul at vt.edu
Sat May 14 11:40:05 CEST 2016



On 5/14/16 3:48 AM, Husen R wrote:
> Hi,
>
> Currently I'm running this tutorial (
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/gmx-tutorials/lysozyme/08_MD.html)
> to simulate restart with fewer nodes.
> at restart, I changed the amount of nodes from 3 to 2 nodes.
> I also changed the amount of processes from 24 to 16 processes.
>
> While the application is running, I tried to see the output file.
> This is the content of the output file :
>
> #########################output file########################
>
>
> Reading checkpoint file md_0_1.cpt generated: Sat May 14 13:10:25 2016
>
>   #ranks mismatch,
>     current program: 16
>     checkpoint file: 24
>
>   #PME-ranks mismatch,
>     current program: -1
>     checkpoint file: 6
>
> GROMACS patchlevel, binary or parallel settings differ from previous run.
> Continuation is exact, but not guaranteed to be binary identical.
>
> Using 16 MPI processes
> Using 1 OpenMP thread per MPI process
>
> starting mdrun 'LYSOZYME in water'
> 500000 steps,   1000.0 ps (continuing from step 54500,    109.0 ps).
>
> ############################################################################
>
> I got a mismatch note as described in the output file above. it is not a
> problem, isn't it ?
> I just want to make sure.
>

This is the message I mentioned in my first reply.  It just means you're now 
changing the DD configuration, PME nodes, etc. so it's not binary identical, but 
the state is faithfully preserved.

http://www.gromacs.org/Documentation/How-tos/Extending_Simulations#Exact_vs_binary_identical_continuation

> is it not allowed to use a different user when we restart simulation from
> checkpoint file ?
> Previously, I failed to restart simulation based on checkpoint file. I
> guess, it is failed because I used a different user (Only a guess).

Presumably one just needs correct read/write permissions, though I have never 
tried to switch users when doing a continuation.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================


More information about the gromacs.org_gmx-users mailing list