[QE-users] occupation=smearing and magnetic moment

[Evgeny Permyakov]

Dear QE users! I found that the calculated magnetic moment of the minimal
cell of LaFe4P12  depends on whether I use occupations=smearing,
specifically I get magnetisation about 2.7 Bohr/cell for occupations=fixed
and zero magnetisation for occupations=smearing. Is it normal? Why does it
even happen? What result is the more 'correct' one?

The system is suspected to be a Pauli paramagnet, does this fact mess
somehow the calculation and how should I approach modeling it?

I added the minimal input file to reproduce the behaviour.


La.Fe4.P12.up.inp
Description: chemical/gamess-input
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] occupation=smearing and magnetic moment

[mkondrin]

On 15.04.2021 15:23, Evgeny Permyakov wrote:

Dear QE users! I found that the calculated magnetic moment of the minimal
cell of LaFe4P12  depends on whether I use occupations=smearing,
specifically I get magnetisation about 2.7 Bohr/cell for occupations=fixed
and zero magnetisation for occupations=smearing. Is it normal? Why does it
even happen? What result is the more 'correct' one?

The system is suspected to be a Pauli paramagnet, does this fact mess
somehow the calculation and how should I approach modeling it?

I added the minimal input file to reproduce the behaviour.



___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Hi, Evgeny!

This can happen if you choose large smearing parameter. So the smearing 
is larger than the energy splitting induced by magnetization. Try to 
decrease degauss value.


Yours,
M. V. Kondrin
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

[QE-users] Help on QE + GPU + OpenMP compilation

[Giovanni Cantele]

Hi all,I’m trying to compile the latest QE version with OpenMP and GPU support. The latest NVIDIA hpc package is in use (nvidia/hpc_sdk/Linux_x86_64/21.3). 4xTesla V100 SXM2 32GB are mounted on the computing node.There are some points that are not clear to me and I would be grateful if someone might help me in this respect:i) configure commandexport PATH=/opt/nvidia/hpc_sdk/Linux_x86_64/21.3/comm_libs/openmpi/openmpi-3.1.5/bin/:/opt/nvidia/hpc_sdk/Linux_x86_64/21.3/compilers/bin/:$PATH./install/configure FC=nvfortran CC=nvcc --enable-openmp --with-cuda=/usr/local/cuda --with-cuda-cc=70 --with-cuda-runtime=11.2 LIBDIRS=/opt/nvidia/hpc_sdk/Linux_x86_64/21.3/compilers/lib/ /opt/nvidia/hpc_sdk/Linux_x86_64/21.3/math_libs/11.2/targets/x86_64-linux/lib/stubs /opt/nvidia/hpc_sdk/Linux_x86_64/21.3/comm_libs/openmpi/openmpi-3.1.5/lib/ /opt/nvidia/hpc_sdk/Linux_x86_64/21.3/comm_libs/mpi/lib/I’ve given the needed paths and libraries that should suffice to a proper installation with CUDA, lapack, blas and so on. The following compilation does work. However, I find something strange inmake.inc :LAPACK_LIBS    = -L/opt/nvidia/hpc_sdk/Linux_x86_64/21.3/comm_libs/openmpi/openmpi-3.1.5/lib/ -llapack  -lblasThe -L…. directory, among those provided through LIBDIRS, is NOT the one containing lapack and blas libraries, so although the subsequent compilation seems to work, the link as above seems to be wrong (it shouldn’t work!). Is that normal? For example, it seems that a library containing dgeem is not found. However, /opt/nvidia/hpc_sdk/Linux_x86_64/21.3/compilers/lib/  contains both libblas.a and liblapack.a .The install/config.log is attached for further checks.ii) CUDA support seems to have been correctly included. I tried to run pw.x and it seems to work. Also, using nvidia-smi, I see that the pw.x processes are running on the GPUs. However, while on Marconi100 the output file contains a section like    GPU acceleration is ACTIVE.     GPU used by master process:        Device Number: 0        Device name: Tesla V100-SXM2-16GB        Compute capability : 70        Ratio of single to double precision performance  : 2        Memory Clock Rate (KHz): 877000        Memory Bus Width (bits): 4096        Peak Memory Bandwidth (GB/s): 898.05, I do not find any message like this in my output. Does it mean that the GPU support has some troubles?iii) last point, it seems that OpenMP support has not been included. I cannot find any indication of that either in config.log or in make.inc (also attached to this message), despite —enable-openmp was specified to the configure command. I also do not find any information in the output concerning the number of threads. Is there anything else I should specify at compiling / running time ?I’m very sorry for overburdening you with so many questions, but any help would be very much appreciated.Thanks,    Giovanni

config.log
Description: Binary data


make.inc
Description: Binary data

-- Giovanni Cantele, PhDCNR-SPINc/o Dipartimento di FisicaUniversita' di Napoli "Federico II"Complesso Universitario M. S. Angelo - Ed. 6Via Cintia, I-80126, Napoli, Italye-mail: giovanni.cant...@spin.cnr.itPhone: +39 081 676910Skype contact: giocan74ResearcherID: http://www.researcherid.com/rid/A-1951-2009Web page: https://sites.google.com/view/giovanni-cantele/home

___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] Help on QE + GPU + OpenMP compilation

[Paolo Giannozzi]

On Thu, Apr 15, 2021 at 2:01 PM Giovanni Cantele <
giovanni.cant...@spin.cnr.it> wrote:

I’m trying to compile the latest QE version
>

"latest stable" or "latest development"? GPU support is work in progress
(or maybe "in regress"): don't count on the stable version to support the
latest compiler.

I can tell you for sure that with a correct nvidia hpc_sdk 21.3
installation + all needed environment variables defined + all needed
configure options (+ MKL libraries: I don't trust too much blas and lapack
from the nvidia compiler) the latest development version compiles and works.

Paolo
-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] Help on QE + GPU + OpenMP compilation

[Giovanni Cantele]

It was the lates stable (qe-6.7-ReleasePack.tgz).

Where can I find the all needed environment variables?

Thanks,

Giovanni

> On 15 Apr 2021, at 14:44, Paolo Giannozzi  wrote:
> 
> On Thu, Apr 15, 2021 at 2:01 PM Giovanni Cantele 
> mailto:giovanni.cant...@spin.cnr.it>> wrote:
> 
> I’m trying to compile the latest QE version
> 
> "latest stable" or "latest development"? GPU support is work in progress (or 
> maybe "in regress"): don't count on the stable version to support the latest 
> compiler. 
>  
> I can tell you for sure that with a correct nvidia hpc_sdk 21.3 installation 
> + all needed environment variables defined + all needed configure options (+ 
> MKL libraries: I don't trust too much blas and lapack from the nvidia 
> compiler) the latest development version compiles and works.
> 
> Paolo
> -- 
> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
> Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
> 
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users

___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] Help on QE + GPU + OpenMP compilation

[Giovanni Cantele]

I downloaded the latest development version. Could you share the needed 
configure options and environment variables that allowed you to get a clean 
compilation?

Thank you again,

Giovanni

> On 15 Apr 2021, at 14:50, Giovanni Cantele  
> wrote:
> 
> It was the lates stable (qe-6.7-ReleasePack.tgz).
> 
> Where can I find the all needed environment variables?
> 
> Thanks,
> 
> Giovanni
> 
>> On 15 Apr 2021, at 14:44, Paolo Giannozzi > > wrote:
>> 
>> On Thu, Apr 15, 2021 at 2:01 PM Giovanni Cantele 
>> mailto:giovanni.cant...@spin.cnr.it>> wrote:
>> 
>> I’m trying to compile the latest QE version
>> 
>> "latest stable" or "latest development"? GPU support is work in progress (or 
>> maybe "in regress"): don't count on the stable version to support the latest 
>> compiler. 
>>  
>> I can tell you for sure that with a correct nvidia hpc_sdk 21.3 installation 
>> + all needed environment variables defined + all needed configure options (+ 
>> MKL libraries: I don't trust too much blas and lapack from the nvidia 
>> compiler) the latest development version compiles and works.
>> 
>> Paolo
>> -- 
>> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>> Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
>> Phone +39-0432-558216, fax +39-0432-558222
>> 
>> ___
>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu 
>> )
>> users mailing list users@lists.quantum-espresso.org 
>> 
>> https://lists.quantum-espresso.org/mailman/listinfo/users 
>> 
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users

___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] occupation=smearing and magnetic moment

[Evgeny Permyakov]

Interesting. With degauss = 1.0e-2 I get zero magnetisation, with degauss =
3.1e-3 I get 0.5 Bohr magnetisation. That's a quite small value, IMHO. Is
it normal?

On Thu, 15 Apr 2021 at 14:41, mkondrin  wrote:

> On 15.04.2021 15:23, Evgeny Permyakov wrote:
> > Dear QE users! I found that the calculated magnetic moment of the minimal
> > cell of LaFe4P12  depends on whether I use occupations=smearing,
> > specifically I get magnetisation about 2.7 Bohr/cell for
> occupations=fixed
> > and zero magnetisation for occupations=smearing. Is it normal? Why does
> it
> > even happen? What result is the more 'correct' one?
> >
> > The system is suspected to be a Pauli paramagnet, does this fact mess
> > somehow the calculation and how should I approach modeling it?
> >
> > I added the minimal input file to reproduce the behaviour.
> >
> >
> >
> > ___
> > Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> > users mailing list users@lists.quantum-espresso.org
> > https://lists.quantum-espresso.org/mailman/listinfo/users
> Hi, Evgeny!
>
> This can happen if you choose large smearing parameter. So the smearing
> is larger than the energy splitting induced by magnetization. Try to
> decrease degauss value.
>
> Yours,
> M. V. Kondrin
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] Help on QE + GPU + OpenMP compilation

[Paolo Giannozzi]

This is what I have used, with some success. If you do not have MKL
libraries, you should get
   BLAS_LIBS=-llapack -lblas
or something like that

Paolo

On Thu, Apr 15, 2021 at 5:27 PM Giovanni Cantele <
giovanni.cant...@spin.cnr.it> wrote:

> I downloaded the latest development version. Could you share the needed
> configure options and environment variables that allowed you to get a clean
> compilation?
>
> Thank you again,
>
> Giovanni
>
> On 15 Apr 2021, at 14:50, Giovanni Cantele 
> wrote:
>
> It was the lates stable (qe-6.7-ReleasePack.tgz).
>
> Where can I find the all needed environment variables?
>
> Thanks,
>
> Giovanni
>
> On 15 Apr 2021, at 14:44, Paolo Giannozzi  wrote:
>
> On Thu, Apr 15, 2021 at 2:01 PM Giovanni Cantele <
> giovanni.cant...@spin.cnr.it> wrote:
>
> I’m trying to compile the latest QE version
>>
>
> "latest stable" or "latest development"? GPU support is work in progress
> (or maybe "in regress"): don't count on the stable version to support the
> latest compiler.
>
> I can tell you for sure that with a correct nvidia hpc_sdk 21.3
> installation + all needed environment variables defined + all needed
> configure options (+ MKL libraries: I don't trust too much blas and lapack
> from the nvidia compiler) the latest development version compiles and works.
>
> Paolo
> --
> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
> Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
>
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
>
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users



-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
# run this file (e.g. "cuda.sh") as ". ./cuda.sh", then:
# ./configure --enable-openmp  --with-cuda=$CUDA_HOME --with-cuda-cc=XX 
--with-cuda-runtime=11.0 --without-scalapack
# where XX depends upon the model of GPU: 35,60,70, or whatever makes you win 
the lottery
export VERSION=21.3
export CUDA_VERSION=11.0

export NVHPC=/opt/nvidia/hpc_sdk

nvcudadir=$NVHPC/Linux_x86_64/$VERSION/cuda
nvcompdir=$NVHPC/Linux_x86_64/$VERSION/compilers
nvmathdir=$NVHPC/Linux_x86_64/$VERSION/math_libs
nvcommdir=$NVHPC/Linux_x86_64/$VERSION/comm_libs

export OPAL_PREFIX=$nvcommdir/mpi

export MANPATH=$MANPATH:$nvcompdir/man

PATH=$nvcudadir/bin:$nvcompdir/bin:$nvcommdir/mpi/bin:$PATH
export PATH

CPATH=$nvcudadir/include:$nvmathdir/include:$nvcommdir/mpi/include:$nvcommdir/nccl/include:$nvcommdir/nvshmem/include:$CPATH
export CPATH

export MKLROOT=/opt/intel/compilers_and_libraries_2020.3.275/linux/mkl
LD_LIBRARY_PATH=$nvcudadir/lib64:$nvcompdir/lib:$nvmathdir/lib64:$nvcommdir/mpi/lib:$nvcommdir/nccl/lib:$nvcommdir/nvshmeme/lib:$MKLROOT/lib/intel64/:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH

export OMPI_FC=nvfortran
export CUDA_HOME=$nvcudadir/$CUDA_VERSION

___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users