[DFTB-Plus-User] large-scale calculations

Ben Hourahine benjamin.hourahine at strath.ac.uk
Fri May 8 16:52:24 CEST 2020


Hello Hafid,

In your script, can you try replacing

dftb+

with

mpirun dftb+

as you seem to have the binary started only on one process.

At the moment you are using 1 processor per k-point, which depending on
system size may not be enough. for ELSI solvers you may have to change
the number of k-points and processor groups, as you should use at least
as many groups as k-points for those solvers (and hence that number of
processors). The default scalapack solvers allow multiple k-points in a
group if you do not have sufficiently many processors.

Regards

Ben


On 08/05/2020 15:38, khalfoun hafid via DFTB-Plus-User wrote:
> Dear Ben ,
> I try using  DFTB+ with the  ELPA ( mode = 2) solver
> (this is a part of my dftb_in.hsd input file)
> ....
> ....
> Hamiltonian = DFTB {
>   solver = ELPA {
>   mode = 2
> }
> ....
> .....
>   KPointsAndWeights = SupercellFolding {
>     1 0 0
>     0 1 0
>     0 0 38
>     0.0 0.0 0.0
>   }
> .....
> .....
>   Parallel {
>   Groups =4
>    Blacs {
>      BlockSize=64
> }
>  }
>
> Howover , after the submission of this script
>  
> This is the script file
> #!/bin/bash
> # Submission script for Lemaitre3
> #SBATCH --job-name=essai
> #SBATCH --time=01:00:00 # hh:mm:ss
> #
> #SBATCH --ntasks=4
> #SBATCH --nodes=1
> #SBATCH --mem-per-cpu=500 # megabytes
> #SBATCH --partition=batch,debug
> #
> #SBATCH --mail-user=h_khalfoun at yahoo.com <mailto:h_khalfoun at yahoo.com>
> #SBATCH --mail-type=ALL
> #ulimit -s unlimited
>
> module load releases/2019b
> module load DFTB+/19.1-intel-2019b-Python-2.7.16
> dftb+
>
>
> I get  this ERROR message:
> --- >  in the run.sh.o (job ID) file :
>  
> |===============================================================================
> |
> |  DFTB+ release 19.1
> |
> |  Copyright (C) 2006 - 2019  DFTB+ developers group
> |
> |===============================================================================
> |
> |  When publishing results obtained with DFTB+, please cite the following
> |  reference:
> |
> |  * B. Aradi, B. Hourahine and T. Frauenheim,
> |    DFTB+, a Sparse Matrix-Based Implementation of the DFTB Method,
> |    J. Phys. Chem. A, 111 5678 (2007).  [doi: 10.1021/jp070186p]
> |
> |  You should also cite additional publications crediting the
> parametrization
> |  data you use. Please consult the documentation of the SK-files for the
> |  references.
> |
> |===============================================================================
>
> Reading input file 'dftb_in.hsd'
> Parser version: 7
>
> --------------------------------------------------------------------------------
> ***  Converting input from version  6 to version  7 ...
> ***  Done.
>
> Reading SK-files:
> Done.
>
>
> Processed input in HSD format written to 'dftb_pin.hsd'
>
> Starting initialization...
> --------------------------------------------------------------------------------
> *ERROR!
> -> Number of groups (4) not compatible with number of processes (1)*
>
> and  this comment in
> in the run.e (job ID ) file 
>
> The following have been reloaded with a version change:
>   1) releases/2018b => releases/2019b
>
> *application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
> [unset]: write_line error; fd=-1 buf=:cmd=abort exitcode=1
> :
> system msg for write_line failure : Bad file descriptor*
> *
> *
>
> I have tried some combination with the MPI parameters.
> Can you tell me what is wrong , please ?
> thanks
> Hafid
>
>
>
> Le mardi 7 avril 2020 à 16:30:27 UTC+2, Ben Hourahine via
> DFTB-Plus-User <dftb-plus-user at mailman.zfn.uni-bremen.de> a écrit :
>
>
> Hello Hafid,
>
> unless there is something special about the electronic structure of
> your system. then the ELPA solver is probably the best choice for
> around 1E3 - 1E4 atoms. This will requires a binary compiled with MPI
> parallelism and both the ELSI and scalapack libraries.
>
> Regards
>
> Ben
>
> On 07/04/2020 15:19, khalfoun hafid via DFTB-Plus-User wrote:
>> Dear Ben ,
>> I would like to calculate band structure of carbon-based structure
>> and plot de the DOS . 
>> By large-scale , I mean structures with great number of atoms ( .i.e.
>> about thousands of atoms )
>>
>> I have MPI in my computational resourses .
>>
>> /Thanks/
>> /Hafid
>> /
>>
>>
>> Le mardi 7 avril 2020 à 16:03:15 UTC+2, Ben Hourahine via
>> DFTB-Plus-User <dftb-plus-user at mailman.zfn.uni-bremen.de>
>> <mailto:dftb-plus-user at mailman.zfn.uni-bremen.de> a écrit :
>>
>>
>> Hello Hafid,
>>
>> Can you supply some more information (and what do you mean by
>> large-scale?):
>>
>> 1) What are your computational resources (serial/shared memory or MPI
>> parallel)?
>>
>> 2) Do you have a GPU?
>>
>> 3) Is your system suitable for low order scaling parallel solvers
>> (HOMO-LUMO gap or low spatial dimension)?
>>
>> Regards
>>
>> Ben
>>
>> On 07/04/2020 09:45, khalfoun hafid via DFTB-Plus-User wrote:
>> > Dear Users
>> > Which solver is suitable for large-scale calculations ?
>> >
>> > thanks
>> >
>> > Hafid
>>
>>
>> -- 
>>         Dr.  B.  Hourahine,  Senior  lecturer
>>             SUPA, Department  of  Physics,
>>             University  of  Strathclyde,
>>               John  Anderson  Building,
>>           107 Rottenrow, Glasgow G4 0NG, UK.                   
>>   +44 141 548 2325, benjamin.hourahine at strath.ac.uk
>> <mailto:benjamin.hourahine at strath.ac.uk>
>>
>> The  Department is  a partner  in SUPA,  the Scottish
>>             Universities Physics Alliance
>>
>> The University  of Strathclyde  is a  charitable body,
>>       registered in Scotland, number SC015263
>>
>>
>> _______________________________________________
>> 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://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.zfn.uni-bremen.de%2Fcgi-bin%2Fmailman%2Flistinfo%2Fdftb-plus-user&data=02%7C01%7Cbenjamin.hourahine%40strath.ac.uk%7Cdbf456e55b5541d9c58808d7f35d9205%7C631e0763153347eba5cd0457bee5944e%7C0%7C0%7C637245455516784692&sdata=LdHY8HWlKjDM%2BV9pywzKhlkf5Y6pOexbJ2zcpOCTitY%3D&reserved=0>
>>
>>
>>
>> _______________________________________________
>> 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://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.zfn.uni-bremen.de%2Fcgi-bin%2Fmailman%2Flistinfo%2Fdftb-plus-user&data=02%7C01%7Cbenjamin.hourahine%40strath.ac.uk%7Cb1530a5d2ac546c818f508d7dafec03d%7C631e0763153347eba5cd0457bee5944e%7C0%7C0%7C637218659998601768&sdata=9qVVNUMua2%2BfqlQT3Vzd6OSviitAdzeAC4eXfvKzR6E%3D&reserved=0 <https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.zfn.uni-bremen.de%2Fcgi-bin%2Fmailman%2Flistinfo%2Fdftb-plus-user&data=02%7C01%7Cbenjamin.hourahine%40strath.ac.uk%7Cdbf456e55b5541d9c58808d7f35d9205%7C631e0763153347eba5cd0457bee5944e%7C0%7C0%7C637245455516794687&sdata=Mwixl62w2aNEGRrsFm3VwA5KRN7efkQnHDgfJvdUZ9c%3D&reserved=0>
> -- 
>         Dr.  B.  Hourahine,  Senior  lecturer
>             SUPA, Department  of  Physics,
>              University  of  Strathclyde,
>               John  Anderson  Building,
>           107 Rottenrow, Glasgow G4 0NG, UK.                    
>   +44 141 548 2325, benjamin.hourahine at strath.ac.uk <mailto:benjamin.hourahine at strath.ac.uk>
>
> The  Department is  a partner  in SUPA,  the Scottish
>             Universities Physics Alliance
>
> The University  of Strathclyde  is a  charitable body,
>        registered in Scotland, number SC015263
> _______________________________________________
> 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://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.zfn.uni-bremen.de%2Fcgi-bin%2Fmailman%2Flistinfo%2Fdftb-plus-user&data=02%7C01%7Cbenjamin.hourahine%40strath.ac.uk%7Cdbf456e55b5541d9c58808d7f35d9205%7C631e0763153347eba5cd0457bee5944e%7C0%7C0%7C637245455516804682&sdata=Atj8Wn0N5TcFjbpR0ariAEYqnu%2FNsc0uJrpFaLbcEM8%3D&reserved=0>
>
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at mailman.zfn.uni-bremen.de
> https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.zfn.uni-bremen.de%2Fcgi-bin%2Fmailman%2Flistinfo%2Fdftb-plus-user&data=02%7C01%7Cbenjamin.hourahine%40strath.ac.uk%7Cdbf456e55b5541d9c58808d7f35d9205%7C631e0763153347eba5cd0457bee5944e%7C0%7C0%7C637245455516824670&sdata=4yqUIh4Oew0UDm0IhLpJ6o1um7o9g3yFhUYTlCZvUkE%3D&reserved=0

-- 
        Dr.  B.  Hourahine,  Senior  lecturer
            SUPA, Department  of  Physics,
             University  of  Strathclyde,
              John  Anderson  Building,
          107 Rottenrow, Glasgow G4 0NG, UK.                    
  +44 141 548 2325, benjamin.hourahine at strath.ac.uk

The  Department is  a partner  in SUPA,  the Scottish
            Universities Physics Alliance

The University  of Strathclyde  is a  charitable body,
       registered in Scotland, number SC015263

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20200508/d2cd9ea5/attachment-0001.htm>


More information about the DFTB-Plus-User mailing list