[DFTB-Plus-User] error with ELSI solvers

Bálint Aradi aradi at uni-bremen.de
Mon Feb 7 22:01:08 CET 2022


Dear Huy,

you probably obtained the non-mpi version (which is the default on 
Conda, if you do not ask for mpi explicitely).

My suggestion would be to create a separate conda environment (e.g. 
dftbplus-mpich or dftbplus-openmpi) for the mpi version, to activate 
that environment and then to download the mpi package into that environment:

conda create -n dftbplus-mpich
conda activate dftbplus-mpich
conda install 'dftbplus=21.2=mpi_mpich_*'

(or the same with openmpi instead). If you afterwards execute

conda list

to investigate the packages within this environment, you should see 
(among others)

elsi                      2.8.2           mpi_mpich_h6ba1590_2 
conda-forge
dftbplus                  21.2            mpi_mpich_h3469437_2 
conda-forge

showing that the mpi-version of dftb+ and also elsi were automatically 
installed.

Best regards,

Bálint


On 06.02.22 07:30, Huy Pham wrote:
> Dear Bálint,
> 
> Thanks for your suggestion!
> 
> I have tried with the conda version, but I got an error: "Not compiled 
> with ELSI supported solvers".
> I am not sure how to get the `correct` one that was compiled with ELSI.
> 
> Best,
> Huy
> 
> On Tue, Feb 1, 2022 at 9:13 AM Bálint Aradi <aradi at uni-bremen.de 
> <mailto:aradi at uni-bremen.de>> wrote:
> 
>     Dear Huy,
> 
>     I have some problems to reproduce the segfault. I've tried your example
>     with the Conda openmpi version, as well as with an Intel-compiled 21.2.
>     Both were able to cope with your example and went without any problems
>     beyond the first total energy calculations (after 14 scc iterations).
> 
>     Could you eventually also try the Conda version, just to exclude, that
>     it is not a problem with the building process somehow?
> 
>     Best regards,
> 
>     Bálint
> 
> 
> 
>     On 28.01.22 15:57, Huy Pham wrote:
>      > Dear Bálint,
>      >
>      > Thanks for your quick reply!
>      >
>      > I am able to solve the PEXSI problem following your suggestion. But
>      > again after the first SCF, there is an error "segmentation fault ".
>      >
>      > Adding "ulimit -s unlimited" before running the exe file doesn't
>     help,
>      > still the same error. Please let me know if you have further
>      > suggestions. Attached are my input files, if you need it.
>      >
>      > Thanks,
>      > Huy
>      >
>      > On Fri, Jan 28, 2022 at 12:12 AM Bálint Aradi
>     <aradi at uni-bremen.de <mailto:aradi at uni-bremen.de>
>      > <mailto:aradi at uni-bremen.de <mailto:aradi at uni-bremen.de>>> wrote:
>      >
>      >     Dear Huy,
>      >
>      >     For the segfault: Often a limited stack size is the reason
>     (hits me
>      >     often). Could you check, whether that is the case? (ulimit -s
>     unlimited)
>      >
>      >     As for the PEXSI problem. The error message is generated by
>     the PEXSI
>      >     library within ELSI. I am definitely not a PEXSI expert, but
>     having a
>      >     quick look at the PEXSI code in ELSI
>      >
>      >
>     https://gitlab.com/ElectronicStructureLibrary/elsi-interface/-/blob/master/src/elsi_util.f90#L423
>     <https://gitlab.com/ElectronicStructureLibrary/elsi-interface/-/blob/master/src/elsi_util.f90#L423>
>      >   
>       <https://gitlab.com/ElectronicStructureLibrary/elsi-interface/-/blob/master/src/elsi_util.f90#L423 <https://gitlab.com/ElectronicStructureLibrary/elsi-interface/-/blob/master/src/elsi_util.f90#L423>>
>      >
>      >     suggests, that the product of the number of poles per process
>      >     (ProcsPerPole), the number of poles (Poles) and the number of
>      >     processors
>      >     used to search the Fermi level (muPoints) should be greater
>     or equal
>      >     than the number of the MPI processes.
>      >
>      >     I hope this helps.
>      >
>      >     Best regards,
>      >
>      >     BálintBest regards,
>      >
>      >     Bálint
>      >
>      >
>      >
>      >     On 28.01.22 01:58, Huy Pham wrote:
>      >      > Hi All,
>      >      >
>      >      > I have a problem when running short MD simulation using
>     DFTB plus
>      >     with
>      >      > different solvers in ELSI. I used the 3ob dataset, the
>     system is a
>      >      > molecular crystal (C, H, N, O) with 576 atoms (supercell).
>      >      >
>      >      > The simulation was fine with ELPA, but if I changed the
>     solver to
>      >     OMM or
>      >      > NTPoly, the simulations crashed after finishing the first SCF.
>      >     The error
>      >      > is "SIGSEGV, segmentation fault occurred" which is
>     difficult to
>      >     track down.
>      >      >
>      >      > When using PEXSI, the error is below:
>      >      > **Error! MPI task      38 in elsi_dm_complex_sparse: Number of
>      >     MPI tasks
>      >      > per pole too small
>      >      > **Error! MPI task     270 in elsi_dm_complex_sparse: Number of
>      >     MPI tasks
>      >      > per pole too small
>      >      > **Error! MPI task       3 in elsi_dm_complex_sparse: Number of
>      >     MPI tasks
>      >      > per pole too small
>      >      > **Error! MPI task     219 in elsi_dm_complex_sparse: Number of
>      >     MPI tasks
>      >      > per pole too small
>      >      > I couldn't find any related info in the manual. According
>     to the
>      >     error
>      >      > message, I have tried to increase the number of MPI tasks,
>     still
>      >     waiting
>      >      > for the simulation to run. Reducing the number of poles
>     doesn't
>      >     help, as
>      >      > at some point, the error "Too few PEXSI poles" will happen.
>      >      >
>      >      > Any suggestion would be appreciated!
>      >      >
>      >      > Thanks,
>      >      > Huy
>      >      >
>      >      > --
>      >      >
>      >      > C. Huy Pham, Ph.D.
>      >      >
>      >      > Staff Scientist
>      >      >
>      >      > Lawrence Livermore National Laboratory
>      >      >
>      >      > Physical and Life Sciences Directorate
>      >      >
>      >      > Materials Science Division, L-287
>      >      >
>      >      > 7000 East Ave, Livermore, CA 94550
>      >      >
>      >      > Phone: (925) 422-7881
>      >      >
>      >      > _______________________________________________
>      >      > DFTB-Plus-User mailing list
>      >      > DFTB-Plus-User at mailman.zfn.uni-bremen.de
>     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
>      >     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de
>     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>>
>      >      >
>      >
>     https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>     <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user>
>      >   
>       <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user>>
>      >
>      >
>      >     --
>      >     Dr. Bálint Aradi
>      >     Bremen Center for Computational Materials Science, University
>     of Bremen
>      > http://www.bccms.uni-bremen.de/cms/people/b-aradi/
>     <http://www.bccms.uni-bremen.de/cms/people/b-aradi/>
>      >     <http://www.bccms.uni-bremen.de/cms/people/b-aradi/
>     <http://www.bccms.uni-bremen.de/cms/people/b-aradi/>>
>      >
>      >     _______________________________________________
>      >     DFTB-Plus-User mailing list
>      > DFTB-Plus-User at mailman.zfn.uni-bremen.de
>     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
>      >     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de
>     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>>
>      >
>     https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>     <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user>
>      >   
>       <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user>>
>      >
>      >
>      >
>      > --
>      >
>      > C. Huy Pham, Ph.D.
>      >
>      > Staff Scientist
>      >
>      > Lawrence Livermore National Laboratory
>      >
>      > Physical and Life Sciences Directorate
>      >
>      > Materials Science Division, L-287
>      >
>      > 7000 East Ave, Livermore, CA 94550
>      >
>      > Phone: (925) 422-7881
>      >
>      > _______________________________________________
>      > DFTB-Plus-User mailing list
>      > DFTB-Plus-User at mailman.zfn.uni-bremen.de
>     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
>      >
>     https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>     <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user>
> 
> 
>     -- 
>     Dr. Bálint Aradi
>     Bremen Center for Computational Materials Science, University of Bremen
>     http://www.bccms.uni-bremen.de/cms/people/b-aradi/
>     <http://www.bccms.uni-bremen.de/cms/people/b-aradi/>
> 
>     _______________________________________________
>     DFTB-Plus-User mailing list
>     DFTB-Plus-User at mailman.zfn.uni-bremen.de
>     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
>     https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>     <https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user>
> 
> 
> 
> -- 
> 
> C. Huy Pham, Ph.D.
> 
> Staff Scientist
> 
> Lawrence Livermore National Laboratory
> 
> Physical and Life Sciences Directorate
> 
> Materials Science Division, L-287
> 
> 7000 East Ave, Livermore, CA 94550
> 
> Phone: (925) 422-7881
> 
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at mailman.zfn.uni-bremen.de
> https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user


-- 
Dr. Bálint Aradi
Bremen Center for Computational Materials Science, University of Bremen
http://www.bccms.uni-bremen.de/cms/people/b-aradi/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 840 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20220207/0e8da937/attachment.sig>


More information about the DFTB-Plus-User mailing list