「From SageMaker Studio」 — 「Manage Amazon SageMaker HyperPod Spaces」 — AWS brings HyperPod Spaces management into SageMaker Studio

On October 6, 2026, AWS announced that users can manage Amazon SageMaker Spaces on Amazon SageMaker HyperPod EKS clusters directly through SageMaker Studio. The interface supports creating, configuring, starting, stopping, and opening Spaces, and lets users launch JupyterLab and Code Editor environments without command-line tools.

The controls are in the IDE and Notebooks tab on a HyperPod cluster detail page. A searchable Spaces table shows each environment’s application type, status, access type, storage, and GPU and vCPU allocations. Users can start or stop a Space with one selection, open it in a browser, or connect through a remote IDE.

The change brings routine Space operations into a graphical interface. AWS says those tasks previously relied primarily on HyperPod CLI or kubectl; its documentation also describes command-line workflows for managing Spaces. HyperPod provides infrastructure for foundation-model training and inference at scale, including distributed training across hundreds of accelerators with resiliency and automatic fault recovery.

Before using the Studio interface, administrators must install the Spaces add-on on the HyperPod EKS cluster and configure access. AWS says custom installation is required for browser access. Configuring the add-on itself carries no additional charge, but users pay for HyperPod compute consumed by Spaces; remote IDE connectivity through SSH-over-SSM also incurs an hourly Systems Manager Advanced On-Premises Instance charge.