[gmx-users] Restart Replica Exchange
gmx3 at hotmail.com
Sat Sep 4 11:46:56 CEST 2010
In gromacs 4.0 the checkpoint files of the different replicas might not be in sync (written at the same step).
In that case you will have to use tpbconv with the trajectory of each replica and select the last common frame.
In the 4.5 release I fixed this issue by synchronizing the checkpointing.
From: mark.abraham at anu.edu.au
To: gmx-users at gromacs.org
Date: Fri, 3 Sep 2010 09:31:38 +1000
Subject: Re: [gmx-users] Restart Replica Exchange
----- Original Message -----
From: Nimesh Jain <nimeshjain2010 at u.northwestern.edu>
Date: Friday, September 3, 2010 1:13
Subject: [gmx-users] Restart Replica Exchange
To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> I was wondering if anyone knows whether it is possible to restart a replica exchange simulation if it stops during the run. I tried it once before and got the error that all the replicas should be at the same time step for replica exchange. When I checked the time step of different replicas, there was a few 10s of picoseconds difference b/w them. Is there a way to circumvent this problem and restart replica exchange?
This issue will normally only arise if your run finished abnormally. I seem to recall discussion of it somewhere. Perhaps you need to restart in a different manner (hint: telling us your command lines would have been useful), e.g. using the checkpoint files properly. Alternatively, since the differences cannot span a replica exchange, you could try running the lagging simulations in serial until they're all in sync.
gmx-users mailing list gmx-users at gromacs.org
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
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the gromacs.org_gmx-users