Launching a DLAMI instance - AWS Deep Learning AMIs

Launching a DLAMI instance

After you find the ID of the DLAMI that you want to use to launch a DLAMI instance, you're ready to launch the instance. To launch it, you can use either the Amazon EC2 console or the AWS Command Line Interface (AWS CLI).

Note

For this walkthrough, we might make references specific to the Deep Learning Base OSS Nvidia Driver GPU AMI (Ubuntu 22.04). Even if you select a different DLAMI, you should be able to follow this guide.

EC2 console
Note

To accelerate high-performance computing (HPC) and machine learning applications, you can launch your DLAMI instance with an Elastic Fabric Adapter (EFA). For specific instructions, see Launching a AWS Deep Learning AMIs Instance With EFA.

  1. Open the EC2 Console.

  2. Note your current AWS Region in the top-most navigation. If this isn't your desired Region, then change this option before you continue. For more information, see Amazon EC2 service endpoints in the Amazon Web Services General Reference.

  3. Choose Launch Instance.

  4. Enter a name for your instance and select the DLAMI that is right for you.

    1. Find an existing DLAMI in My AMIs or choose Quick Start.

    2. Search by DLAMI ID. Browse the options then select your choice.

  5. Choose an instance type. You can find the recommended instance families for your DLAMI in Deep Learning AMIs Release Notes. For general recommendations on DLAMI instance types, see Choosing a DLAMI instance type.

  6. Choose Launch Instance.

AWS CLI

After you launch your instance using either the Amazon EC2 console or AWS CLI, wait for the instance to be ready. This usually takes only a few minutes. You can verify the status of the instance in the Amazon EC2 console. For more information, see Status checks for Amazon EC2 instances in the Amazon EC2 User Guide.

Next step

Connecting to a DLAMI instance