Computer Lab: Difference between revisions

From cmpb
Jump to navigationJump to search
Akefayati (talk | contribs)
Akefayati (talk | contribs)
Line 21: Line 21:
==OpenMX input files==
==OpenMX input files==
*[https://t-ozaki.issp.u-tokyo.ac.jp/lectures-2023/OpenMX-2.pdf Structure of OpenMX input files explained]
*[https://t-ozaki.issp.u-tokyo.ac.jp/lectures-2023/OpenMX-2.pdf Structure of OpenMX input files explained]
*Silicon using OpenMX: [[Media:Hands-On.pdf|PDF]][[Media:Si.zip|files]]
*Silicon using OpenMX: [[Media:Hands-On.pdf|PDF]], [[Media:Si.zip|files]]
*Magnons in Fe using OpenMX: [[Media:Hands-On-Fe.pdf|PDF]][[Media:Fe.zip|files]]
*Magnons in Fe using OpenMX: [[Media:Hands-On-Fe.pdf|PDF]], [[Media:Fe.zip|files]]
*Density of states of NiO via DFT vs. DFT+ U: [[Media:Hands-On-NiO.pdf|PDF]][[Media:NiO.zip|files]]
*Density of states of NiO via DFT vs. DFT+ U: [[Media:Hands-On-NiO.pdf|PDF]], [[Media:NiO.zip|files]]
*[https://pranabdas.github.io/openmx/convergence Testing convergence in kinetic energy cutoff, k-grid density and lattice constant]
*[https://pranabdas.github.io/openmx/convergence Testing convergence in kinetic energy cutoff, k-grid density and lattice constant]

Revision as of 10:14, 19 August 2024

Basic Unix

$ module load vasp/5.4.4-intel-2023
$ mpirun -n 4 vasp_std
  • How to submit parallel openmx jobs over 16 cores of keldysh:
$ module load openmx
$ mpirun -np 16 openmx material.dat -nt 4 > material.std

Note that OpenMX has hybrid MPI/OpenMP parallelization, so nt specifies the number of threads in each process managed by MPI. If '-nt' is not specified, then the number of threads is set to 1, which corresponds to the flat MPI parallelization.

VASP input files

OpenMX input files