Provision and manage SageMaker HyperPod clusters on EKS and Slurm.
npx skills add https://github.com/dgallitelli/aws-hyperpod-skill --skill sagemaker-hyperpod
Please help me install this Agent Skill. Skill: sagemaker-hyperpod Source: https://github.com/dgallitelli/aws-hyperpod-skill/tree/main Command: npx skills add https://github.com/dgallitelli/aws-hyperpod-skill --skill sagemaker-hyperpod
HyperPod cluster provisioning and management for distributed ML training in SageMaker, enabling teams to deploy scalable GPU and Trainium-based clusters quickly and with governance.
Install the HyperPod CLI, run hyp init to create a cluster stack, validate the configuration with hyp validate, and deploy the cluster with hyp create.
High-intent search queries and answers about installing and using this skill.
To provision a SageMaker HyperPod cluster, use the HyperPod CLI to initialize a stack with hyp init, validate the configuration with hyp validate, and deploy the cluster using hyp create for GPU and Trainium workloads.
Yes, SageMaker HyperPod supports both EKS and Slurm orchestrators for cluster creation, job submission, and monitoring across distributed training workloads on multi-node GPU and Trainium environments.
Before creating a HyperPod cluster, you must pass prerequisites validation, verify AWS service quotas, integrate lifecycle scripts, and ensure add-on compatibility to maintain reliable and compliant deployments.
You can submit and monitor distributed training jobs on SageMaker HyperPod by utilizing the integrated EKS or Slurm orchestrator commands provided by the Skill for multi-node production-grade workloads.
To troubleshoot a SageMaker HyperPod deployment, use the built-in troubleshooting guidance to validate prerequisites, check service quotas, and verify lifecycle script integration for production-grade workloads.
Yes, SageMaker HyperPod provisions and manages distributed ML training clusters equipped with both Trainium and GPU accelerators for scalable production-grade multi-node workloads.
haozhx23Automate EKS cluster creation, dependency configuration, and HyperPod provisioning.
haozhx23Diagnose NCCL failures on SageMaker HyperPod clusters via events, security groups, and node status.
gabrielfruetSubmit and monitor Slurm benchmark jobs on remote HPC clusters.
haozhx23Diagnose and remediate SageMaker HyperPod cluster issues with AWS tooling.
