GPU - Get Started

Deploy a GPU Linode Instance

1. Log in to the Cloud Manager with the username and password you created when signing up.

2. At the top of the page, click Create and select Linode.

3. Select the Distribution, Marketplace App, or Image you would like to use.

   Note

   Be sure to select a distribution that’s compatible with the NVIDIA CUDA Toolkit. Review NVIDIA’s System Requirements to learn which distributions are supported.

4. Choose the region where you would like your Linode to reside. If you’re not sure which to select, see our How to Choose a Data Center guide. You can also generate MTR reports for a deeper look at the route path between you and a data center in each specific region.

5. Select a GPU Linode plan.

6. Give your Linode a label. This is a name to help you easily identify it within the Cloud Manager’s Dashboard. If desired, assign a tag to the Linode in the Add Tags field.

7. Create a root password for your Linode in the Root Password field. This password must be provided when you log in to your Linode via SSH. The password must meet the complexity strength validation requirements for a strong password. Your root password can be used to perform any action on your server, so make it long, complex, and unique.

8. Click Create. You are directed back to the Linode’s page, and this page reports the status of your Linode as it boots up. You can now use the Cloud Manager to:

   • Boot and shut down your Linode

   • Access monitoring statistics

   • Update your billing and account information

   • Add additional Linode services, like Block Storage

   • Open a support ticket and perform other administrative tasks

9. Be sure to bookmark the Linode Status page or subscribe to our system status updates by email.

10. After the Linode GPU is online, install the NVIDIA CUDA Toolkit:

    • Installing the NVIDIA CUDA Toolkit