Introducing new Ray capabilities on SageMaker HyperPod
Amazon brings managed Ray to SageMaker HyperPod—native Kubernetes support for distributed training without the ops overhead.
Why it matters
SageMaker HyperPod's new Ray integration lets ML engineers run distributed training and inference on Kubernetes without managing Ray infrastructure separately. This consolidates the dev experience for practitioners building multi-node workloads on AWS.
The key facts
11 to knowAWS SageMaker HyperPod adds managed Ray support on Amazon EKS
Ray clusters can be created and monitored from SageMaker Studio
JupyterLab and Code Editor notebooks can connect to live Ray clusters
Built on open-source KubeRay and standard Ray APIs
Includes out-of-the-box observability for distributed training and inference
Resilient distributed training capability
SageMaker HyperPod adds managed Ray support on Amazon EKS
Built-in observability for Ray clusters
Supports resilient distributed training and accelerated inference
Uses open-source KubeRay and standard Ray APIs
Launches from SageMaker Studio
Go to the source
AWS Machine Learning Blogaws.amazon.com
Publisher excerpt: Amazon SageMaker HyperPod now offers managed Ray support on Amazon EKS. Create and monitor Ray clusters, connect JupyterLab and Code Editor notebooks to live clusters, get out-of-the-box observability, and run resilient distributed training and accelerated inference from SageMaker Studio, all with…