Amazon SageMaker HyperPod adds managed extensions for Slurm clusters

Covered by 2 sources
Read onAWS Blog· 21 min
Amazon SageMaker HyperPod has launched managed cluster extensions for Slurm clusters that enable API-driven configuration and continuous provisioning of node software. The initial features focus on observability and user management, improving cluster software management without replacing nodes.

What happened

Amazon Web Services has launched managed cluster extensions for Amazon SageMaker HyperPod Slurm clusters. These new extensions allow administrators to configure and update node software through APIs or the SageMaker console without needing to replace nodes. The initial rollout includes features for observability and user management, aiming to streamline software management on running clusters. The update maintains direct node access for administrators, so manual configurations and scripts remain possible alongside the new automated tools.

Key details

  • The managed cluster extensions launched in November 2026.
  • Administrators can now manage node software using APIs or the SageMaker AI console.
  • Initial extensions include observability (metrics collection) and user management (directory integration).
  • Cluster software is kept consistent across all nodes, including new or replaced ones.
  • Direct node access and custom lifecycle scripts are still supported.

Why it matters

Platform teams and administrators using SageMaker HyperPod Slurm clusters can now manage software and settings across all nodes more efficiently, reducing manual maintenance and the risk of configuration drift. This change allows for quicker updates and consistent software environments without taking nodes offline or replacing them. Organizations benefit from improved cluster reliability and reduced operational overhead while retaining the flexibility to customize nodes directly if needed.

Also covered by 1 source

More on this story

  1. Introducing enhanced infrastructure observability in Amazon SageMaker HyperPod
    · AWS Blog
  2. Introducing workload recovery and expanded infrastructure checks on Amazon SageMaker HyperPod
    · 2 sources
  3. Detecting and recovering from silent data corruption on Amazon SageMaker HyperPod
    · AWS Blog

More in Industry & Manufacturing