Run OpenFOAM docker on a SLURM cluster

Here’s a comprehensive guide on running OpenFOAM within Docker containers on a SLURM cluster, along with best practices and troubleshooting tips:

1. Prerequisites

• SLURM Cluster: A functioning SLURM cluster with access to compute nodes.
• Docker Engine: Docker Engine must be installed and configured on each compute node within your SLURM cluster.
• OpenFOAM Image: A suitable Docker image containing your desired OpenFOAM version, libraries, and dependencies. You can build a custom image or use a pre-built one from Docker Hub (e.g., openfoam/openfoam-latest).
• Your OpenFOAM Case: The directory containing your OpenFOAM simulation case files (e.g., system, constant, 0, etc.).
• SLURM Script: You’ll need a SLURM script to submit your job to the cluster.

2. Creating a Docker Image (if needed)

• Dockerfile: Create a Dockerfile to define your OpenFOAM image. Example:

FROM openfoam/openfoam-latest

WORKDIR /home/openfoam

# Install additional packages (if needed)
RUN apt-get update && apt-get install -y python3 python3-pip

# Copy your case files
COPY . /home/openfoam/my_case

# Define the entrypoint (your OpenFOAM command)
ENTRYPOINT ["/opt/openfoam/bin/foamRun", "simpleFoam"]

• Build the image:docker build -t my-openfoam-image .

3. Writing the SLURM Script

#!/bin/bash
#SBATCH --job-name=my_openfoam_job
#SBATCH --nodes=1
#SBATCH --ntasks-per-node=1
#SBATCH --cpus-per-task=4
#SBATCH --mem=8G
#SBATCH --time=0-10:00
#SBATCH --output=my_openfoam_job-%j.out
#SBATCH --error=my_openfoam_job-%j.err

# Load Docker environment
module load docker

# Run the Docker container 
docker run -it -v $(pwd)/my_case:/home/openfoam/my_case my-openfoam-image

Explanation of the Script:

• Header: This section specifies job details (name, resources, time, etc.).
• Load Docker: Loads the Docker module on your cluster.
• docker run: Starts a Docker container.
  • -it: Runs interactively and attaches the terminal.
  • -v: Mounts your case directory inside the container. This makes your case files available within the container.
  • my-openfoam-image: The name of your Docker image.

4. Submitting the Job

sbatch my_openfoam_script.sh

Important Considerations:

• Mount Points: Ensure you are correctly mounting your case directory inside the container to give OpenFOAM access to the files it needs.
• Resources: Carefully allocate resources like CPU cores, memory, and time. Overestimation can waste resources, while underestimation can lead to job failure.
• Image Size: Keep your Docker image as small as possible to reduce transfer times and storage overhead.
• Parallelism: Consider how to effectively parallelize your simulations within the Docker container.
• Network Access: If your simulations require network access (e.g., for external data sources or visualization tools), you’ll need to configure your container to allow network access.
• Cleaning Up: Consider automating the cleanup of the container and any generated data after your job completes.

Example with Pre-built Image:

You can use a pre-built OpenFOAM image without creating your own:

#!/bin/bash
#SBATCH ... (as above)

module load docker

# Use the openfoam/openfoam-latest image
docker run -it -v $(pwd)/my_case:/home/openfoam/my_case openfoam/openfoam-latest 

Troubleshooting:

• Check Logs: Inspect the output files (my_openfoam_job-%j.out and my_openfoam_job-%j.err) to troubleshoot problems.
• SLURM Errors: Look for error messages from SLURM itself.
• Docker Errors: Check the Docker logs for problems starting the container or executing commands.

Advanced Techniques:

• Singularity: If you’re looking for a more lightweight alternative to Docker, Singularity can be a good choice.
• MPI: Use MPI within your OpenFOAM simulations for parallel computing.
• GPU Acceleration: Utilize GPUs for faster simulations by using specialized OpenFOAM images with GPU support.

Let me know if you have any further questions or specific requirements for your OpenFOAM setup!

---

CloudHPC is a HPC provider to run engineering simulations on the cloud. CloudHPC provides from 1 to 224 vCPUs for each process in several configuration of HPC infrastructure - both multi-thread and multi-core. Current software ranges includes several CAE, CFD, FEA, FEM software among which OpenFOAM, FDS, Blender and several others.

New users benefit of a FREE trial of 300 vCPU/Hours to be used on the platform in order to test the platform, all each features and verify if it is suitable for their needs

---

Contact Name/Surname *

Company Name *

Company address *

City *

State/Province *

Country *CountryAfghanistanAlbaniaAlgeriaAmericaAndorraAngolaAntiguaArgentinaArmeniaAustraliaAustriaAzerbaijanBahamasBahrainBangladeshBarbadosBelarusBelgiumBelizeBeninBhutanBissauBoliviaBosniaBotswanaBrazilBritishBruneiBulgariaBurkinaBurmaBurundiCambodiaCameroonCanadaCape VerdeCentral African RepublicChadChileChinaColombiaComorosCongoCosta RicaCroatiaCubaCyprusCzechDenmarkDjiboutiDominicaEast TimorEcuadorEgyptEl SalvadorEmirateEnglandEritreaEstoniaEthiopiaFijiFinlandFranceGabonGambiaGeorgiaGermanyGhanaGreat BritainGreeceGrenadaGrenadinesGuatemalaGuineaGuyanaHaitiHerzegovinaHollandHondurasHungaryIcelandIndiaIndonesiaIranIraqIrelandIsraelItalyIvory CoastJamaicaJapanJordanKazakhstanKenyaKiribatiKoreaKosovoKuwaitKyrgyzstanLaosLatviaLebanonLesothoLiberiaLibyaLiechtensteinLithuaniaLuxembourgMacedoniaMadagascarMalawiMalaysiaMaldivesMaliMaltaMarshallMauritaniaMauritiusMexicoMicronesiaMoldovaMonacoMongoliaMontenegroMoroccoMozambiqueMyanmarNamibiaNauruNepalNetherlandsNew ZealandNicaraguaNigerNigeriaNorwayOmanPakistanPalauPanamaPapuaParaguayPeruPhilippinesPolandPortugalQatarRomaniaRussiaRwandaSamoaSan MarinoSao TomeSaudi ArabiaScotlandScottishSenegalSerbiaSeychellesSierra LeoneSingaporeSlovakiaSloveniaSolomonSomaliaSouth AfricaSouth SudanSpainSri LankaSt KittsSt LuciaSudanSurinameSwazilandSwedenSwitzerlandSyriaTaiwanTajikistanTanzaniaThailandTobagoTogoTongaTrinidadTunisiaTurkeyTurkmenistanTuvaluUgandaUkraineUnited KingdomUnited StatesUruguayUSAUzbekistanVanuatuVaticanVenezuelaVietnamWalesWelshYemenZambiaZimbabwe

Email *

Email

Confirm Email

Privacy *

• I accept the privacy conditions

Privacy Policy

Message

Send registration