Hi,
Another issue that I notice it is crucial is that the number of computed
frames has to be multiple of the number of threads used in parallel
computing otherwise the calculation finishes abnormally. This is my
experience.
Campa,
2015-10-27 13:29 GMT+01:00 Josep Maria Campanera Alsina <campaxic.gmail.com>
:
> Thanks Tyler and Jason for your replies,
>
> I've followed your advice. First I've run with a standard prmtop and the
> problem persisted. Parallel and serial MMPBSA.py script gave rise to the
> same problem. The last lines of _MMPBSA_complex_rism.mdout.0 do not show
> any thing abnormal. Finally I've checked the memory issue and indeed this
> was the origin of the problem. Polardecomp=1 requires significant amount of
> memory. So, problem solved!
>
> Thank you very much,
>
> Campa,
>
>
> 2015-10-22 21:33 GMT+02:00 Josep Maria Campanera Alsina <
> campaxic.gmail.com>:
>
>> Dear all,
>> I get the following error when trying to decompose (polardecomp=1) the
>> solvation free energy from 3D-RISM in MMPBSA.py. The calculation actually
>> does a couple of frames and finally crashes.
>>
>>
>> ----------------------------------------------------------------------------------------
>> Beginning 3D-RISM calculations with
>> /Users/campa/Amber/amber14/bin/rism3d.snglpnt
>> calculating complex contribution...
>> File "/Users/campa/Amber/amber14/bin/MMPBSA.py.MPI", line 96, in
>> <module>
>> app.run_mmpbsa()
>> File "/Users/campa/Amber/amber14/bin/MMPBSA_mods/main.py", line 218, in
>> run_mmpbsa
>> self.calc_list.run(rank, self.stdout)
>> File "/Users/campa/Amber/amber14/bin/MMPBSA_mods/calculation.py", line
>> 79, in run
>> calc.run(rank, stdout=stdout, stderr=stderr)
>> File "/Users/campa/Amber/amber14/bin/MMPBSA_mods/calculation.py", line
>> 269, in run
>> Calculation.run(self, rank, stdout=self.output % rank)
>> File "/Users/campa/Amber/amber14/bin/MMPBSA_mods/calculation.py", line
>> 148, in run
>> self.prmtop))
>> CalcError: /Users/campa/Amber/amber14/bin/rism3d.snglpnt failed with
>> prmtop AB40.WT-WT.nowat.ildn.pdb.charmm.pdb.psfgen.pdb.amber.top!
>> Error occured on rank 0.
>> Exiting. All files have been retained.
>> application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
>>
>> ----------------------------------------------------------------------------------------
>>
>> with the simple following input:
>>
>> ------------------------------------------
>> &general
>> startframe=1,endframe=5000,interval=500,verbose=2, keep_files=2,
>> netcdf=1, full_traj=1, use_sander=0,
>> /
>> &rism
>> thermo="gf",polardecomp=1,rism_verbose=2,
>> /
>> ------------------------------------------
>>
>> The error appears either in serial or parallel calculations in MMPBSA.py
>> script of AMBER14. I use the ff99SB-ILDN forcefield and the topology file
>> was created originally with chamber but finally converted to "nornal" amber
>> topology file with cpptraj facility with the option "parmwrite out ...
>> nochamber". The same calculation without the option "polardecomp=1" ends
>> correctly. As it seems a problem related with settings of radii (CalcError)
>> I've tried to change the radii using parmed but the problem persists.
>>
>> All comments welcome and thank you very much for the help,
>>
>> Campa,
>>
>
>
>
> --
> ---------------------------------------------
> Josep Maria Campanera Alsina
> Professor Agregat Interí
> Departament de Fisicoquímica
> Facultat de Farmàcia
> Universitat de Barcelona
> Avgda Joan XXIII, s/n
> 08028 Barcelona
> Tel: +34 93 4035985
> Fax: +34 93 4035987
> campanera.ub.edu
> http://campanerablog.wordpress.com <https://campanerablog.wordpress.com/>
> http://science.lsi.upc.edu
> --------------------------------------------
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Nov 12 2015 - 05:00:03 PST
