Hi Jmao,
I generally do the same (restart every X hours, accordingly to the batch scheduler specifications. If you use Slurm (but you can modify it to make it work with other schedulers), I can share with you a script that automatically does the restart every N steps (you can tune N accordingly to your resources). You can restart as many times as you need, provided that you specify irest=1, ntx=5 in the input (which is generally already set, since you start the production after an equilibration phase).
Unfortunately I do not understand the second question...
Best,
Alessandro
<signaturebeforequotedtext></signaturebeforequotedtext>
Il 17/06/20 09:32, 毛君 <junmaozb.foxmail.com> ha scritto:
>
> Hello everyone! There are some problems when I do REMD simulation and I need help！
> Firstly，I must do the simulation restart every 72h due t o cluster settings the task just run no more than three days, otherwise it will be killed。I don’t know whether frequent restarting tasks will affect the accuracy of the REMD simulation, and what should I pay attention to when restarting? Can I still use the original input file？
> Secondly， it possible to use a similar but different PDB file after clustering as an input file for a REMD simulation?
> These two problems troublesome to me and I hope you can help!
> Thank you for your attention to this matter.
>
> jmao
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
<signatureafterquotedtext>--
Prof. Alessandro Contini, PhD
DISFARM - Sezione di Chimica Generale e Organica "Alessandro Marchesini"
Università degli Studi di Milano
Via Venezian, 21 20133 Milano
Tel. +390250314480 Fax. +390250314476
e-mail alessandro.contini.unimi.it
skype alessandrocontini</signatureafterquotedtext>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jun 17 2020 - 01:00:04 PDT
