Ansys Fluent on Qarnot Cloud

Ansys Fluent is a powerful computational fluid dynamics (CFD) software that simulates fluid flow, heat transfer, and other related phenomena. It is widely utilized in various sectors, including aerospace, automotive, energy, and environmental engineering.

Researchers, engineers, and designers rely on Fluent to analyze complex fluid behavior, optimize designs, and improve performance. The software employs numerical methods to solve partial differential equations governing fluid flow, providing accurate predictions and insights for real-world applications.

License

In order to launch a Fluent simulation on Qarnot, you must grant Qarnot access to your license. For more details, please reach out to our team at support-computing@qarnot-computing.com.

We will assume your license setup with us is already complete from this point on.

Versions

Here are the current available versions on Qarnot:

• 2025 R1 - GPU available
• 2024 R2 - GPU available
• 2024 R1 - GPU available
• 2023 R2 - GPU available
• 2023 R1 - GPU available
• 2022 R1
• 2021 R2                          

If you are interested in another version, please send us an email support-computing@qarnot-computing.com.

Prerequisites

Before launching a computation task, please ensure that the following prerequisites have been met:

• Create an account
• Retrieve your authentication token
• Install one of Qarnot’s SDK (Python) (Node.js) (C#) (Commande Line)
• Know your Qarnot profiles, e.g. fluent-e-corp and fluent-e-corp-ssh

Launching an MPI test

Ansys-Fluent mpitest can run without a license. To start a MPI test on Qarnot, copy the following code in a Python script. Be sure you have copied your authentication token in the script (instead of MY_SECRET_TOKEN) to be able to launch the task on Qarnot.

‍

Good to know :  when you create a task with conn.create_task, you have to specify its name, the profile you want to use and in a cluster case, the number of instances you want run.

To run this script, open a terminal in your working directory and execute python3 run_fluent_mpi-test.py. It will launch the MPI test on Qarnot.

Launching a use-case

Here is an example of how to run an Ansys-Fluent use-case on the platform. It assumes that

• you have access to a license.
• your use-case includes data and a .jou file in an input/ folder in your working directory.

If needed, here is the generic use case aicraft_wing_14m. You can download it and extract it inside an input/ directory.

Once you have all setup, it's time to execute run_fluent_aircraft-wing.py below. In this script you can

• Select the directory you want to synchronize your bucket with
• Indicates the FLUENT_CMD you want use
• Specify your license server IP/Port.

To launch the computation on Qarnot, copy the following code in a Python script in your working directory. Be sure you have copied your authentication token in the script (instead of MY_SECRET_TOKEN) and execute python3 run_fluent_aircraft_wing.py.

Launching a Fluent task with SSH connectivity enabled

Here is an example of how to run an Ansys-Fluent with SSH activated on the platform. You will need:

• An SSH public key (you can create one by following this tutorial)
• An access to a license to execute the Fluent commands you wan

To launch the computation on Qarnot, copy the following code in a Python script in your working directory. Be sure you have:

• copied your authentication token in the script (instead of MY_SECRET_TOKEN)
• copied your SSH Key in DOCKER_SSH

And now, you can execute python3 run_fluent_ssh.py.

Please note the profile used has changed, its now ansys-fluent-ssh.

While connected to the master node, you can launch a simulation on your cluster with this command:

fluent 3ddp -g mpi=openmpi -pinfiniband -cnf=/job/mpihosts -i run.jou

Where mpi=openmpi -pinfiniband -cnf=/job/mpihosts are MPI settings needed for more than 1 node simulation.

Monitoring

At any given time, you can monitor the status of your task on Tasq.

On this page you can monitor your instances states, logs or execution time and check your task configuration you set up such as a profile name, the Fluent version or your SSH key.

In case of an SSH task, IP and port can be found under the Port Forward section.

Your results will be stored in the fluent-aircraft-output bucket and can be retrieved with three methods:

• the download button in the Bucket section
• the download_results function from the Python SDK
• use s3 compatible opensource tools

For a large quantity of files or more than 5GB of data, it is recommended to use rclone (Linux) or Cyberduck (Windows). These are two open-source applications for S3 bucket management.

Wrapping up

If you have any questions, please contact support-computing@qarnot-computing.com and we will help you with pleasure!