Quantum-espresso has been using MPI long before openmp, as a result the
MPI parallelism is more developped and more stable. I think you can
stick to just MPI with good result, even on a single workstation.
hth
On 15/10/17 08:11, Rajesh wrote:
Dear Prof Paulatto ,
Than you for your response. I need to use qe for openmpi as well
openmp but on separate workstations. I think openmp works better on
multi-cpu pc and openmpi on multi node clusters. I am right?
Thank you.
On Sun, Oct 15, 2017 at 12:42 AM, Lorenzo Paulatto <paul...@gmail.com
<mailto:paul...@gmail.com>> wrote:
On 14/10/17 09:15, Rajesh wrote:
> Dear Prof Giannozzi,
> Thank you for your reply. But I need openmpi parallelization.
openmpi is NOT OpenMP, the former is an open-source implementation of
MPI, the later is shared-memory vectorization based on compile-time
instructions. The first is required to run in parallel, you need both
for huge number of CPUs.
What do you need exactly ?
--
Lorenzo Paulatto - Paris
_______________________________________________
Pw_forum mailing list
Pw_forum@pwscf.org <mailto:Pw_forum@pwscf.org>
http://pwscf.org/mailman/listinfo/pw_forum
<http://pwscf.org/mailman/listinfo/pw_forum>
_______________________________________________
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum
--
Lorenzo Paulatto - Paris
_______________________________________________
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum