Skip to main content

Setting up your cluster following the best practices

Initial cluster setup​

To create a SageMaker HyperPod in just a few clicks, navigate to the Amazon Sagemaker AI console and click on HyperPod Clusters. Under Cluster Management, click on the Create HyperPod Cluster button. More details, read the official AWS documentation here.

Amazon Sagemaker Hyperpod cluster creation experience

  • With the new quick setup path, you can now launch a fully-operational cluster optimized for large-scale AI workloads directly from the AWS console using a streamlined single-page interface that provisions all dependencies including VPCs, subnets, FSx storage, EKS orchestrator, and essential configurations required for building and deploying models.

  • The custom setup path empowers platform engineering teams already familiar with AWS to fine-tune every setting, from specific subnet configurations to selective installations, all within the same unified console experience, along with the ability to export auto-generated CloudFormation templates for production deployment.

To get started, navigate to the SageMaker AI console and follow the instructions at: Getting started with SageMaker HyperPod using the SageMaker console

Console experience with a single instance group

After you click on Submit, you will see your cluster being created. You can check the console to verify what's the status of this process. When the clsuter shows as InService then you can start using it. The whole process usually don't take more than 20 minutes to be ready.