AI Tools - Remote SSH Connection

Objective

This tutorial covers the process of starting a job using a Visual Studio Code Remote via SSH.

Requirements

• An AI Notebook or AI Training Project created inside a Public Cloud project in your OVHcloud account
• An AI user
• The OVHcloud AI CLI installed on your computer

OVHcloud Console Access

• Direct link: Projects
• Navigation path: Public Cloud > Select your project > AI > AI Notebooks

Instructions

Installation

1. Install an OpenSSH compatible SSH client if one is not already present.
2. Install Visual Studio Code.
3. Install the Remote Development extension pack.

Generate an SSH Keypair

If you don’t already have an SSH key, you will need to create one. To do so, open a Terminal / PowerShell, and run the key generation command:

ssh-keygen -t ed25519 -C "your_email@example.com"

This will create a private key (e.g., id_ed25519, not to share) and a public key (e.g., id_ed25519.pub) in your ~/.ssh/ directory.

Specify the SSH Key during AI Notebook or AI Training Job Creation

Here is how to add your SSH key to your AI Solution if you are using the OVHcloud Console or the ovhai CLI:

ovhai job run <job_image> \
  --gpu <nb-gpus> \
  --ssh-public-keys ~/.ssh/id_ed25519.pub

ovhai job get <job-id>

Once your job is created, regardless of the method chosen, you will be able to copy your job id (e.g., bfa1d77a-9746-4128-9974-f94139937927), which will be needed in the following steps.

Verify you can connect to the SSH host

Before continuing, you can verify you can access your job by running the following command from a terminal / PowerShell window, replacing <job-id> with the ID of your job and <region> with your resource location (e.g., gra, bhs):

ssh <job-id>@<region>.ai.cloud.ovh.net

If the SSH key is well configured, you should see the following:

Welcome to OVHcloud!
ovh@job-36ed1f18-626b-42f7-b15f-0ed844e65d20:~$

Configure VSCode Remote Connection

In Visual Studio Code, click on the Remote Explorer icon.

From there, click the gear icon on the SSH feature to access the SSH configuration file for VSCode, typically located at ~/.ssh/config, and open it.

Add the following section for your new host:

Host my_ai_training_job
HostName <region>.ai.cloud.ovh.net # Adapt to your resource location (gra, bhs)
User bfa1d77a-9746-4128-9974-f94139937927 # ID of your AI Training Job or AI Notebook
IdentityFile C:\Path\To\id_ed25519 # Path to your private key file (not .pub)

Save the config file. Then refresh remotes by clicking on this icon:

Your resource should now appear in the SSH menu on the left. Connect to it by clicking the arrow next to your resource. VSCode will restart to connect to your remote resource:

If the connection is well established, you should see this message in the bottom-left corner of VSCode:

Develop and Run Your Code

Once connected, you can start developing and running your code in the default /workspace folder, by clicking on the Open folder blue button, and select /workspace.

You can also open a terminal and run commands from it. Enjoy!

Go further

You can compare AI models based on resource consumption, accuracy and training time. Refer to this tutorial.

For training or technical assistance implementing our solutions, contact your sales representative or visit our Professional Services page to request a quote and have your project analysed by our experts.

Feedback

Please send us your questions, feedback and suggestions to improve the service:

• On the OVHcloud Discord server