AMBER Archive (2008)

Subject: Re: AMBER: RE: About restart amber

From: Francesco Pietra (chiendarret_at_gmail.com)
Date: Sat Nov 22 2008 - 10:03:02 CST

Hi Ross:
When restraints are imposed, which reference file should be used in
running sander or pmemd again? I guess the "-ref ...rst" pertaining to
the crashed job. Correct? Does that involve any further complication?

In view of non minor work in restarting from a crashed job while
taking everything into account (not to mention the unluky situation
that the machine was just writing the rst file), is any major drawback
in making the various steps of MD short? If problems exist, how to
determine how much short?

Thanks
francesco pietra

On Fri, Nov 21, 2008 at 10:45 PM, Ross Walker <ross_at_rosswalker.co.uk> wrote:
> Hi Zhou,
>
>
>
> Such questions should really go to the AMBER mailing list which is where I
> am copying this message. (See http://ambermd.org/) for signup details.
>
>
>
> To answer your question, assuming the machine did not crash while writing
> the restart file, you can restart the simulation in the same way that one
> runs a single MD simulations in multiple steps. See for example how it is
> done in section 6 of tutorial B1.
>
>
>
> http://ambermd.org/tutorials/basic/tutorial1/section6.htm
>
>
>
> Essentially you create a new input file with irest=1 and ntx=5 and then run
> sander or pmemd again providing the rst file from the crashed job as the
> input coordinates (-c ...rst). The code will then continue running from the
> point at which the last restart file was written - make sure you write to a
> new output and mdcrd file so you don't overwrite the previous run.
>
>
>
> If you still want the exact same length of calculation then you should look
> in the top of the restart file where the simulation time that the restart
> file corresponds to is written and then adjust nstlim in the new mdin file
> accordingly. You might also need to manually clean up your output file so it
> stops at the point at which the restart file was written (if ntpr /= ntwr).
> The same is true for the mdcrd file although you will need to run this
> through ptraj and have it write you a new mdcrd file containing all the
> frames from 1 to the frame the restart file corresponds to.
>
>
>
> Good luck,
>
> Ross
>
>
>
> From: z g [mailto:zgong.hust_at_gmail.com]
> Sent: Thursday, November 20, 2008 12:50 AM
> To: Ross Walker
> Subject: About restart amber
>
>
>
> Dear sir :
> I have read the online Amber beginning tutorials, and learned a
> lot. Thank you very much!
> Here, I have some questions about using amber
> I put a task on sander, during the simulation, some accident happened ( such
> as power cut ), and my program stopped.
> How can I restart the program and continue doing my simulations. Please tell
> me the details about it
> Thanks for your patience!
>
> your sincerely
> Zhou Gong
>
>
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu