How to create and manage a GPU Instance

Scaleway GPU Instances feature dedicated high-end Nvidia GPUs, making them optimal for tasks such as data processing, artificial intelligence, rendering, and video encoding.

Once you’ve created your GPU Instance, connecting to it via SSH allows you to use our pre-configured Docker images, providing instant access to a preinstalled environment with your preferred AI libraries and tools already set up.

Moreover, GPU Instances offer all the functionalities of our standard Instances, including flexible IPs, Security Groups, Private Networks, backups, and more.

When you have completed your calculations using the GPU Instance, deletion can be done through the Scaleway console, API, or our CLI tools.

Before you start

To complete the actions presented below, you must have:

• A Scaleway account logged into the console
• Owner status or IAM permissions allowing you to perform actions in the intended Organization
• An SSH key added to your account

How to create a GPU Instance

1. Navigate to Instances under the Compute section in the side menu of the Scaleway console. This action will direct you to the Instance dashboard.
2. Click *+ Create Instance to proceed to the Instance creation page.
3. Follow these steps to configure your GPU Instance:
   • Availability Zone: Choose the geographical region, represented by the Availability Zone, where your Instance will be deployed.
   • Instance Type (GPU): Select the desired GPU Instance type, considering factors such as processing power, memory, storage options, and bandwidth. Refer to our guide on choosing the right GPU Instance type for more information.
   • Image: Pick an operating system image suitable for your GPU Instance. For example, select Ubuntu Jammy GPU OS 12, which comes with preinstalled NVIDIA drivers and an NVIDIA Docker environment. You have the flexibility to customize your working environment using Docker with our provided Docker images or your own containers.
   • Volumes: Optionally, add storage volumes for your Instance. You can adjust settings such as Block and Local Storage volumes according to your requirements.
     Note

     • The recommended minimum volume size for GPU OS images is 125 GB.
     • If your GPU Instance supports scratch storage, the scratch volume displays but can not be edited. Learn more about scratch storage.
   • Network Configuration: Choose between a routed public IP or a NAT public IP for your Instance. We recommend using a routed public IP. You can allocate IPv4 and IPv6 addresses as needed, with a maximum of 5 IPs per Instance.
   • Instance Name and Tags: Assign a name to your Instance for easy identification. You can also add tags to organize your Instances efficiently.
   • Advanced Options: Configure cloud-init settings if required, or leave them at default values.
   • SSH Keys: Verify the SSH keys that will grant you access to your Instance.
   • Estimated Cost: Review the estimated cost of your GPU Instance based on the selected specifications.
4. Once you have completed the configuration, click on Create Instance to initiate the creation process. You will receive a notification once the GPU Instance is ready for use.

How to connect to a GPU Instance

See our documentation on how to connect to your Instance via SSH.

Once you have connected via SSH, you can launch a Docker container to start working on your AI projects.

How to use Instance features

For instructions on using any type of GPU Instance feature, including flexible IPs, placement groups, Private Networks, backups, and much more, check out our full Instance how-to documentation.

How to delete a GPU Instance

See our documentation on how to delete your Instance.