[gmx-users] error in the middle of running mdrun_mpi

Justin Lemkul jalemkul at vt.edu
Mon Oct 27 01:16:04 CET 2014 


On 10/26/14 9:55 AM, Nizar Masbukhin wrote:
> regarding gaining speed in implicit solvent simulation, i have tried to
> parallelize using -ntmpi flag. However gromacs doesn't allow as i use group
> cutoff-scheme. Any recommendation how to parallelise implicit solvent
> simulation? I do need parallelise my simulation. I have found the same
> question in this mail list, one suggest use all-vs-all kernel which uses
> zero cut-off.
> This is my test run actually. I intend to run my simulation in cluster
> computer.
>

Unless the restriction was lifted at some point, implicit simulations won't run 
on more than 2 cores.  There were issues with constraints that led to the 
limitation.

-Justin

> On Sun, Oct 26, 2014 at 8:23 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
>
>>
>>
>> On 10/26/14 9:17 AM, Nizar Masbukhin wrote:
>>
>>> Thanks Justin.
>>> I have increased the cutoff, and yeah thats work. There were no error
>>> message anymore. The first 6 nanoseconds, i felt the simulation run
>>> slower.
>>> Felt so curious that  simulation run very fast the rest of time.
>>>
>>>
>> Longer cutoffs mean there are more interactions to calculate, but the
>> cutoffs aren't to be toyed with arbitrarily to gain speed.  They are a
>> critical element of the force field itself, though in implicit solvent, it
>> is common to increase (and never decrease) the cutoff values used in
>> explicit solvent.  Physical validity should trump speed any day.
>>
>> -Justin
>>
>>
>>   On Fri, Oct 24, 2014 at 7:37 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
>>>
>>>
>>>>
>>>> On 10/24/14 8:31 AM, Nizar Masbukhin wrote:
>>>>
>>>>   Thanks for yor reply, Mark.
>>>>>
>>>>>
>>>>> At first i was sure that the problem was table-exension because when I
>>>>> enlarge table-extension value, warning message didn't  appear anymore.
>>>>> Besides, i have successfully minimized and equilibrated the system
>>>>> (indicated by Fmax < emtol reached; and no error messages during NVT&NPT
>>>>> equilibration, except a warning that the Pcouple is turned off in vacuum
>>>>> system).
>>>>>
>>>>> However, the error message appeared without table-extension warning
>>>>> makes
>>>>> me doubt also about my system stability. Here is my mdp setting. Please
>>>>> tell me if there are any 'weird' setting, and also kindly
>>>>> suggest/recommend
>>>>> a better setting.
>>>>>
>>>>>
>>>>> *mdp file for Minimisation*
>>>>>
>>>>>
>>>>> integrator = steep
>>>>>
>>>>> nsteps = 5000
>>>>>
>>>>> emtol = 200
>>>>>
>>>>> emstep = 0.01
>>>>>
>>>>> niter = 20
>>>>>
>>>>> nstlog = 1
>>>>>
>>>>> nstenergy = 1
>>>>>
>>>>> cutoff-scheme = group
>>>>>
>>>>> nstlist = 1
>>>>>
>>>>> ns_type = simple
>>>>>
>>>>> pbc = no
>>>>>
>>>>> rlist = 0.5
>>>>>
>>>>> coulombtype = cut-off
>>>>>
>>>>> rcoulomb = 0.5
>>>>>
>>>>> vdw-type = cut-off
>>>>>
>>>>> rvdw-switch = 0.8
>>>>>
>>>>> rvdw = 0.5
>>>>>
>>>>> DispCorr = no
>>>>>
>>>>> fourierspacing = 0.12
>>>>>
>>>>> pme_order = 6
>>>>>
>>>>> ewald_rtol = 1e-06
>>>>>
>>>>> epsilon_surface = 0
>>>>>
>>>>> optimize_fft = no
>>>>>
>>>>> tcoupl = no
>>>>>
>>>>> pcoupl = no
>>>>>
>>>>> free_energy = yes
>>>>>
>>>>> init_lambda = 0.0
>>>>>
>>>>> delta_lambda = 0
>>>>>
>>>>> foreign_lambda = 0.05
>>>>>
>>>>> sc-alpha = 0.5
>>>>>
>>>>> sc-power = 1.0
>>>>>
>>>>> sc-sigma  = 0.3
>>>>>
>>>>> couple-lambda0 = vdw
>>>>>
>>>>> couple-lambda1 = none
>>>>>
>>>>> couple-intramol = no
>>>>>
>>>>> nstdhdl = 10
>>>>>
>>>>> gen_vel = no
>>>>>
>>>>> constraints = none
>>>>>
>>>>> constraint-algorithm = lincs
>>>>>
>>>>> continuation = no
>>>>>
>>>>> lincs-order  = 12
>>>>>
>>>>> implicit-solvent = GBSA
>>>>>
>>>>> gb-algorithm = still
>>>>>
>>>>> nstgbradii = 1
>>>>>
>>>>> rgbradii = 0.5
>>>>>
>>>>> gb-epsilon-solvent = 80
>>>>>
>>>>> sa-algorithm = Ace-approximation
>>>>>
>>>>> sa-surface-tension = 2.05
>>>>>
>>>>>
>>>>> *mdp file for NVT equilibration*
>>>>>
>>>>>
>>>>> define = -DPOSRES
>>>>>
>>>>> integrator = md
>>>>>
>>>>> tinit = 0
>>>>>
>>>>> dt = 0.002
>>>>>
>>>>> nsteps = 250000
>>>>>
>>>>> init-step = 0
>>>>>
>>>>> comm-mode = angular
>>>>>
>>>>> nstcomm = 100
>>>>>
>>>>> bd-fric = 0
>>>>>
>>>>> ld-seed = -1
>>>>>
>>>>> nstxout = 1000
>>>>>
>>>>> nstvout = 50000
>>>>>
>>>>> nstfout = 50000
>>>>>
>>>>> nstlog = 100
>>>>>
>>>>> nstcalcenergy = 100
>>>>>
>>>>> nstenergy = 1000
>>>>>
>>>>> nstxtcout = 100
>>>>>
>>>>> xtc-precision = 1000
>>>>>
>>>>> xtc-grps = system
>>>>>
>>>>> energygrps = system
>>>>>
>>>>> cutoff-scheme= group
>>>>>
>>>>> nstlist                  = 1
>>>>>
>>>>> ns-type = simple
>>>>>
>>>>> pbc= no
>>>>>
>>>>> rlist= 0.5
>>>>>
>>>>> coulombtype = cut-off
>>>>>
>>>>> rcoulomb= 0.5
>>>>>
>>>>> vdw-type = Cut-off
>>>>>
>>>>> vdw-modifier = Potential-shift-Verlet
>>>>>
>>>>> rvdw-switch= 0.8
>>>>>
>>>>> rvdw = 0.5
>>>>>
>>>>> table-extension = 500
>>>>>
>>>>> fourierspacing = 0.12
>>>>>
>>>>> fourier-nx  = 0
>>>>>
>>>>> fourier-ny = 0
>>>>>
>>>>> fourier-nz = 0
>>>>>
>>>>> implicit-solvent = GBSA
>>>>>
>>>>> gb-algorithm = still
>>>>>
>>>>> nstgbradii = 1
>>>>>
>>>>> rgbradii = 0.5
>>>>>
>>>>> gb-epsilon-solvent = 80
>>>>>
>>>>> sa-algorithm = Ace-approximation
>>>>>
>>>>> sa-surface-tension = 2.05
>>>>>
>>>>> tcoupl = v-rescale
>>>>>
>>>>> nsttcouple = -1
>>>>>
>>>>> nh-chain-length = 10
>>>>>
>>>>> print-nose-hoover-chain-variables = no
>>>>>
>>>>> tc-grps = system
>>>>>
>>>>> tau-t = 0.1
>>>>>
>>>>> ref-t = 298.00
>>>>>
>>>>> pcoupl = No
>>>>>
>>>>> pcoupltype = Isotropic
>>>>>
>>>>> nstpcouple = -1
>>>>>
>>>>> tau-p = 1
>>>>>
>>>>> refcoord-scaling = No
>>>>>
>>>>> gen-vel = yes
>>>>>
>>>>> gen-temp = 298.00
>>>>>
>>>>> gen-seed  = -1
>>>>>
>>>>> constraints= all-bonds
>>>>>
>>>>> constraint-algorithm = Lincs
>>>>>
>>>>> continuation = no
>>>>>
>>>>> Shake-SOR = no
>>>>>
>>>>> shake-tol = 0.0001
>>>>>
>>>>> lincs-order = 4
>>>>>
>>>>> lincs-iter = 1
>>>>>
>>>>> lincs-warnangle = 30
>>>>>
>>>>>
>>>>> *mdp file for NPT equilibration*
>>>>>
>>>>>
>>>>> define = -DPOSRES
>>>>>
>>>>> integrator = md
>>>>>
>>>>> tinit = 0
>>>>>
>>>>> dt = 0.002
>>>>>
>>>>> nsteps = 500000
>>>>>
>>>>> init-step = 0
>>>>>
>>>>> simulation-part = 1
>>>>>
>>>>> comm-mode = angular
>>>>>
>>>>> nstcomm = 100
>>>>>
>>>>> bd-fric = 0
>>>>>
>>>>> ld-seed = -1
>>>>>
>>>>> nstxout = 1000
>>>>>
>>>>> nstvout = 500000
>>>>>
>>>>> nstfout = 500000
>>>>>
>>>>> nstlog = 100
>>>>>
>>>>> nstcalcenergy = 100
>>>>>
>>>>> nstenergy = 1000
>>>>>
>>>>> nstxtcout = 100
>>>>>
>>>>> xtc-precision = 1000
>>>>>
>>>>> xtc-grps = system
>>>>>
>>>>> energygrps = system
>>>>>
>>>>> cutoff-scheme = group
>>>>>
>>>>> nstlist = 1
>>>>>
>>>>> ns-type = simple
>>>>>
>>>>> pbc = no
>>>>>
>>>>> rlist  = 0.5
>>>>>
>>>>> coulombtype= cut-off
>>>>>
>>>>> rcoulomb = 0.5
>>>>>
>>>>> vdw-type = Cut-off
>>>>>
>>>>> vdw-modifier = Potential-shift-Verlet
>>>>>
>>>>> rvdw-switch = 0.8
>>>>>
>>>>> rvdw= 0.5
>>>>>
>>>>> table-extension = 1
>>>>>
>>>>> fourierspacing = 0.12
>>>>>
>>>>> fourier-nx= 0
>>>>>
>>>>> fourier-ny = 0
>>>>>
>>>>> fourier-nz = 0
>>>>>
>>>>> implicit-solvent = GBSA
>>>>>
>>>>> gb-algorithm = still
>>>>>
>>>>> nstgbradii = 1
>>>>>
>>>>> rgbradii = 0.5
>>>>>
>>>>> gb-epsilon-solvent = 80
>>>>>
>>>>> sa-algorithm = Ace-approximation
>>>>>
>>>>> sa-surface-tension = 2.05
>>>>>
>>>>> tcoupl  = Nose-Hoover
>>>>>
>>>>> tc-grps = system
>>>>>
>>>>> tau-t  = 0.1
>>>>>
>>>>> ref-t = 298.00
>>>>>
>>>>> pcoupl = parrinello-rahman
>>>>>
>>>>> pcoupltype = Isotropic
>>>>>
>>>>> tau-p   = 1.0
>>>>>
>>>>> compressibility   = 4.5e-5
>>>>>
>>>>> ref-p   = 1.0
>>>>>
>>>>> refcoord-scaling = No
>>>>>
>>>>> gen-vel   = no
>>>>>
>>>>> gen-temp = 298.00
>>>>>
>>>>> gen-seed   = -1
>>>>>
>>>>> constraints  = all-bonds
>>>>>
>>>>> constraint-algorithm   = Lincs
>>>>>
>>>>> continuation  = yes
>>>>>
>>>>> Shake-SOR  = no
>>>>>
>>>>> shake-tol  = 0.0001
>>>>>
>>>>> lincs-order = 4
>>>>>
>>>>> lincs-iter   = 1
>>>>>
>>>>> lincs-warnangle  = 30
>>>>>
>>>>>
>>>>> *mdp file for MD*
>>>>>
>>>>>
>>>>> integrator  = md
>>>>>
>>>>> tinit = 0
>>>>>
>>>>> dt  = 0.001
>>>>>
>>>>> nsteps = 500000000 ; 1 us
>>>>>
>>>>> init-step = 0
>>>>>
>>>>> simulation-part= 1
>>>>>
>>>>> comm-mode  = Angular
>>>>>
>>>>> nstcomm = 100
>>>>>
>>>>> comm-grps = system
>>>>>
>>>>> bd-fric  = 0
>>>>>
>>>>> ld-seed = -1
>>>>>
>>>>> nstxout  = 10000
>>>>>
>>>>> nstvout  = 0
>>>>>
>>>>> nstfout   = 0
>>>>>
>>>>> nstlog  = 10000
>>>>>
>>>>> nstcalcenergy = 10000
>>>>>
>>>>> nstenergy = 10000
>>>>>
>>>>> nstxtcout  = 0
>>>>>
>>>>> xtc-precision  = 1000
>>>>>
>>>>> xtc-grps  = system
>>>>>
>>>>> energygrps  = system
>>>>>
>>>>> cutoff-scheme  = group
>>>>>
>>>>> nstlist  = 10
>>>>>
>>>>> ns-type  = simple
>>>>>
>>>>> pbc   = no
>>>>>
>>>>> rlist    = 0.5
>>>>>
>>>>> coulombtype    = cut-off
>>>>>
>>>>> rcoulomb  = 0.5
>>>>>
>>>>> vdw-type  = Cut-off
>>>>>
>>>>> vdw-modifier  = Potential-shift-Verlet
>>>>>
>>>>> rvdw-switch = 0.8
>>>>>
>>>>> rvdw  = 0.5
>>>>>
>>>>> DispCorr = No
>>>>>
>>>>> table-extension  = 500
>>>>>
>>>>> fourierspacing = 0.12
>>>>>
>>>>> fourier-nx  = 0
>>>>>
>>>>> fourier-ny = 0
>>>>>
>>>>> fourier-nz = 0
>>>>>
>>>>> implicit-solvent = GBSA
>>>>>
>>>>> ;implicit-solvent = GBSA
>>>>>
>>>>> gb-algorithm = still
>>>>>
>>>>> nstgbradii = 1
>>>>>
>>>>> rgbradii = 0.5
>>>>>
>>>>> gb-epsilon-solvent = 80
>>>>>
>>>>> sa-algorithm = Ace-approximation
>>>>>
>>>>> sa-surface-tension = 2.05
>>>>>
>>>>> tcoupl = v-rescale
>>>>>
>>>>> nsttcouple = -1
>>>>>
>>>>> nh-chain-length = 10
>>>>>
>>>>> print-nose-hoover-chain-variables = no
>>>>>
>>>>> tc-grps = system
>>>>>
>>>>> tau-t  = 0.1
>>>>>
>>>>> ref-t = 298.00
>>>>>
>>>>> pcoupl = No
>>>>>
>>>>> pcoupltype = Isotropic
>>>>>
>>>>> nstpcouple = -1
>>>>>
>>>>> tau-p  = 1
>>>>>
>>>>> refcoord-scaling = No
>>>>>
>>>>> gen-vel = no
>>>>>
>>>>> gen-temp = 298.00
>>>>>
>>>>> gen-seed  = -1
>>>>>
>>>>> constraints  = none
>>>>>
>>>>> constraint-algorithm = Lincs
>>>>>
>>>>> continuation = yes
>>>>>
>>>>> Shake-SOR = no
>>>>>
>>>>> shake-tol = 0.0001
>>>>>
>>>>> lincs-order = 4
>>>>>
>>>>> lincs-iter = 1
>>>>>
>>>>> lincs-warnangle = 30
>>>>>
>>>>> morse = no
>>>>>
>>>>> free-energy = yes
>>>>>
>>>>> couple-lambda0 = vdw-q
>>>>>
>>>>> couple-lambda1= vdw-q
>>>>>
>>>>> couple-intramol = no
>>>>>
>>>>> init-lambda = 0
>>>>>
>>>>> delta-lambda = 0.1
>>>>>
>>>>> nstdhdl  = 50
>>>>>
>>>>> calc-lambda-neighbors  = 1
>>>>>
>>>>> dhdl-print-energy = no
>>>>>
>>>>> sc-alpha = 0
>>>>>
>>>>> sc-power= 1
>>>>>
>>>>> sc-r-power  = 6
>>>>>
>>>>> sc-sigma= 0.3
>>>>>
>>>>> sc-coul = no
>>>>>
>>>>> separate-dhdl-file = yes
>>>>>
>>>>> dhdl-derivatives  = yes
>>>>>
>>>>> dh_hist_size  = 0
>>>>>
>>>>> dh_hist_spacing = 0.1
>>>>>
>>>>>
>>>>>   You need to simplify significantly to determine what the issue is.
>>>> You're
>>>> using (1) very short cutoffs, (2) free energy, and (3) implicit solvent.
>>>> Any combination of those might be causing the problem you're having.
>>>> Generally, extra long cutoffs are used in implicit environments.  Try a
>>>> "normal" run without the free energy code going and see if that improves
>>>> the situation.
>>>>
>>>> -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
>>>>
>>>> ==================================================
>>>>
>>>> --
>>>> Gromacs Users mailing list
>>>>
>>>> * Please search the archive at http://www.gromacs.org/
>>>> Support/Mailing_Lists/GMX-Users_List before posting!
>>>>
>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>
>>>> * For (un)subscribe requests visit
>>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>>> send a mail to gmx-users-request at gromacs.org.
>>>>
>>>>
>>>
>>>
>>>
>> --
>> ==================================================
>>
>> 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
>>
>> ==================================================
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at http://www.gromacs.org/
>> Support/Mailing_Lists/GMX-Users_List before posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to gmx-users-request at gromacs.org.
>>
>
>
>

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

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