-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathjobscript_finerMesh
26 lines (24 loc) · 1003 Bytes
/
jobscript_finerMesh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
#!/bin/bash -l
#SBATCH --partition=standard
## request 1 node; each node offers up to 20 ranks (cpu cores)
#SBATCH --nodes=1
## runtime in hours; after 120h (5 days) our simulations will be stopped
## by the cluster; no problem is the simulation finishes earlier
#SBATCH --time=120:00:00
## name to find your job in the queue
#SBATCH --job-name=plate
## number of requested ranks/cores per node
## total cores = request nodes x ntasks-per-node = 1 x 4 = 4
#SBATCH --ntasks-per-node=4
## load software libraries
module load singularity/3.6.0rc2
module load mpi/openmpi/4.0.1/cuda_aware_gcc_6.3.0
## submit job
echo "Submitting case $1"
cd run/$1
# Add one more variable to choose a turbulence model when executing sbatch
# Currently, 1. kOmegaSST and 2. SpalartAllmaras are available.
# In order to choose a model, we should add one more argument of number in the command line.
# e.g.) "sbatch jobscript_finerMesh turbulentFlatPlate 2" (for SA model)
export model_num=$2
./Allrun_finerMesh.singularity