Learning Forum

[Jayachandran Narayanan]

Hi,

I am trying to run my Fluent simulations in our university HPC cluster and each node of the cluster is limited to 40 processors. So, I have requested 2 nodes and 80 processors to run Fluent cases. The HPC allocated me the resources whereas when I am opening Fluent, it's just loading 80 partitions in the 40 processors of the same node. Fluent console displays 80/40 and not 80/80 at the core column at the starting. 

Please help me in setting up the simulations in multiple nodes. 

[Rahul]

Hello 
Could you please share your job submission script?

[Jayachandran Narayanan]

Hi,

Thanks for your response. 

I am trying to open Fluent using the GUI. Here are settings in Fluent Launcher.

• Solver processors - 80
• Parallel settings: interconnects and MPI types - default, Run type - shared memory on a local machine.

[Rahul]

Please find more information here: 42.2. Starting Parallel Ansys Fluent Using Fluent Launcher

[Jayachandran Narayanan]

Hi, 

I tried different launcher options as in the shared section, but none of them seems to work. Instead, I created a slurm jobscript file as,

#!/bin/sh

#SBATCH --time=300:00:00

#SBATCH -N 2 -n 20

#SBATCH --mem=100gb

module load ansys/22.2

cp 3d_sim.in SYS-3.cas.h5 SYS-3.dat.h5 $PFSDIR

cd $PFSDIR

FLUENTNODES="$(scontrol show hostnames)"

FLUENTNODES=$(echo $FLUENTNODES | tr ' ' ',')

fluent 3ddp -t$SLURM_NTASKS -mpi=openmpi -p -g -t40 -cnf=$FLUENTNODES -i 3d_sim.in

cp * $SLURM_SUBMIT_DIR

It seems like its able to access multiple nodes, but threw this error as,

 

/usr/local/ansys_inc/v222/fluent/fluent22.2.0/bin/fluent -r22.2.0 3ddp -t20 -mpi=openmpi -p -g -t40 -cnf=compt326,compt327 -i 3d_sim.in

/usr/local/ansys_inc/v222/fluent/fluent22.2.0/cortex/lnamd64/cortex.22.2.0 -f fluent -g -i 3d_sim.in (fluent "3ddp -pethernet  -host -r22.2.0 -t40 -mpi=openmpi -cnf=compt326,compt327 -path/usr/local/ansys_inc/v222/fluent -ssh")

 

Opening input/output transcript to file "/scratch/pbsjobs/job.871158.hpc/fluent-20240221-095031-932314.trn".

Auto-Transcript Start Time:  09:50:31, 21 Feb 2024 

/usr/local/ansys_inc/v222/fluent/fluent22.2.0/bin/fluent -r22.2.0 3ddp -pethernet -host -t40 -mpi=openmpi -cnf=compt326,compt327 -path/usr/local/ansys_inc/v222/fluent -ssh -cx compt326:41759:34939

Starting /usr/local/ansys_inc/v222/fluent/fluent22.2.0/lnamd64/3ddp_host/fluent.22.2.0 host -cx compt326:41759:34939 "(list (rpsetvar (QUOTE parallel/function) "fluent 3ddp -flux -node -r22.2.0 -t40 -pethernet -mpi=openmpi -cnf=compt326,compt327 -ssh") (rpsetvar (QUOTE parallel/rhost) "") (rpsetvar (QUOTE parallel/ruser) "") (rpsetvar (QUOTE parallel/nprocs_string) "40") (rpsetvar (QUOTE parallel/auto-spawn?) #t) (rpsetvar (QUOTE parallel/trace-level) 0) (rpsetvar (QUOTE parallel/remote-shell) 1) (rpsetvar (QUOTE parallel/path) "/usr/local/ansys_inc/v222/fluent") (rpsetvar (QUOTE parallel/hostsfile) "compt326,compt327") (rpsetvar (QUOTE gpuapp/devices) ""))"

/usr/local/ansys_inc/v222/fluent/fluent22.2.0/lnamd64/3ddp_host/fluent.22.2.0: error while loading shared libraries: libnsl.so.1: cannot open shared object file: No such file or directory

 The fluent process could not be started.

 

version> 

version> 

version> 

version> 

Can you help us in resolving this?

 

Thanks!