From: Brian Radak (brian.radak_at_gmail.com)
Date: Fri Aug 31 2018 - 09:22:48 CDT
A higher margin seems to have helped your issue some, since the number of
exclusions is much closer to the expected number. Did the simulation also
take longer to return the error?
I generally have good luck with a margin value close to 5.0, but that's
just a random guess. The error message also gives a very nice statement
regarding your chosen values for cutoff and pairlistdist - have you tried
adjust those?
Without knowing anything else about your system, there is very little
further advice that anyone here can give you.
BKR
On Fri, Aug 31, 2018 at 9:40 AM Mani Kandan <manikandand5110_at_gmail.com>
wrote:
> Dear all,
>
> After increasing margin also I am getting the same error.
> Is there any way to resolve these issue.
>
>
> OPENING EXTENDED SYSTEM TRAJECTORY FILE
> FATAL ERROR: Low global exclusion count! (2337723 vs 2337729) System
> unstable or pairlistdist or cutoff too small.
>
> FATAL ERROR: See http://www.ks.uiuc.edu/Research/namd/bugreport.html
>
> Regards,
>
> MANIKANDAN D
>
> On Wed, Aug 29, 2018 at 10:20 PM Radak, Brian K <bradak_at_illinois.edu>
> wrote:
>
>> Please send general use questions to the namd-l list only (not namd)
>>
>> This is a well-established occurrence in NAMD and there are several, if
>> not dozens, of posts on this list regarding the problem and solution.
>>
>> Your density does not appear to be well equilibrated and, as a result,
>> the box is changing volume too quickly for the domain decomposition to keep
>> up. You can:
>>
>> 1) choose a better starting density
>> 2) choose a larger margin (this may hurt performance, so use the default
>> once equilibration is achieved)
>> 3) run shorter NpT segments so that the patches are re-allocated more
>> frequently
>>
>> For very small and very diffuse systems, none of these may work.
>>
>> Brian Radak
>> Postdoctoral Associate, Software Development
>> Theoretical and Computational Biophysics Group
>> Beckman Institute for Advanced Science and Technology
>> University of Illinois at Urbana-Champaign
>> bradak_at_ks.uiuc.edu
>> ------------------------------
>> *From:* Mani Kandan [manikandand5110_at_gmail.com]
>> *Sent:* Wednesday, August 29, 2018 12:35 PM
>> *To:* namd-l_at_ks.uiuc.edu; namd_at_ks.uiuc.edu
>> *Subject:* Error NPT simulation - Need Urgent Help - Reg
>>
>> Dear all,
>> I am getting the following error while running NPT
>> simulation. I am running NPT after minimization, but in minimization, there
>> is no problem. NPT is running for 60 steps and showing this error. The same
>> code runs very well when system dimensions of 100 A. Kindly help me in this
>> regard.
>>
>>
>>
>> The last position output (seq=60) takes 0.198 seconds, 535.676 MB of
>> memory in use
>> WRITING VELOCITIES TO RESTART FILE AT STEP 60
>> FINISHED WRITING RESTART VELOCITIES
>> ERROR: Margin is too small for 25467 atoms during timestep 67.
>> ERROR: Incorrect nonbonded forces and energies may be calculated!
>> FATAL ERROR: High global exclusion count! (1282894 vs 868383) System
>> unstable or pairlistdist or cutoff too close to periodic cell size.
>>
>> FATAL ERROR: See http://www.ks.uiuc.edu/Research/namd/bugreport.html
>>
>>
>> Regards,
>>
>> MANIKANDAN D
>>
>
This archive was generated by hypermail 2.1.6 : Mon Dec 31 2018 - 23:21:22 CST