acusolve issue with slurm

Unknown
Unknown
edited April 2023 in Community Q&A

Hi experts,

I just noticed something in cluster. My submission script to cluster via slurm is:

#########################################################################

#!/bin/bash

#SBATCH --job-name="acupro_concurrent_test"
#SBATCH -o acupro_concurrent_test_%j.out
#SBATCH --ntasks-per-node=36
#SBATCH --ntasks=180
#SBATCH --partition=general

for i in {1..5}
do
    srun -n 1 -c 36 --exclusive acuRun -np 36 -pb acupro_$i -inp half_print.inp -libs ./libusr.so &       ### JUST FOR TEST WITH SAME INPUT FILE
done
wait

#########################################################################

 

The output file is attached herewith. I am looking to run only 5 simulations. It seems like --ntasks divides the same acusolve simulations into 36 tasks. So total 5*36=180 tasks which makes sense for slurm. If you look at the error file, there are exactly 180 lines for all simulations with 36 lines corresponding to each/same simulations. But, I am looking to run only 5 acusolve simulations. 

Do you have any idea why is this happening?

 

Thanks.

Tagged:

Answers

  • acupro
    acupro
    Altair Employee
    edited April 2023

    I would recommend you contact your Slurm support team regarding behavior of that product.

    The reason for acuPrep failing would be more for this platform.

Categories