Page 1 of 1

Error in routine epw_readin (1):

Posted: Sun Jul 09, 2023 11:01 pm
by rkarkee
Hi all,

I am trying to run epw.x but got this error.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine epw_readin (1):
Number of processes must be equal to product of number of pools and
number of images
Image parallelization can be used only in calculations on coarse grid.


I am not using any npools nor image parallelization: Please see my job script below. How can I fix this?
Thank you

#!/bin/bash

#Submit this script with: sbatch filename

#SBATCH --time=16:00:00 # walltime
#SBATCH --nodes=1 # number of nodes
#SBATCH --ntasks-per-node=100 # number of tasks per node
#SBATCH --job-name=epw_HfTe5 # job name
#SBATCH --partition=standard # partition name
#SBATCH --no-requeue # do not requeue when preempted and on node failure
#SBATCH --signal=30@20 # send signal to job at [seconds] before end

module swap PrgEnv-cray PrgEnv-intel
module load cmake


srun -n 100 /usr/projects/icapt/quantum-espresso/q-e/bin/epw.x < epw1.in &> epw1.out

Re: Error in routine epw_readin (1):

Posted: Mon Jul 10, 2023 5:07 pm
by stiwari
Hi,


Please try "srun /usr/projects/icapt/quantum-espresso/q-e/bin/epw.x -nk 100 < epw1.in &> epw1.out". The reason for this error is that the number of cores (100 in your case) should be equal to the number for k-space parallelization (-nk 100). Also, keep in mind that the number of cores you are using should be <= total number of k-points.