- Adds deep health checks for Slurm clusters with continuous provisioning, enabling proactive GPU health validation on new and existing nodes.
- Integrates health check results into console and APIs, isolates unhealthy instances, and works with automatic node recovery.
- Available across all SageMaker HyperPod regions, with documentation updates in the User Guide.
Community impact
Community ratings: 0 Useful, 0 Noise, 0 Risky, 0 Broke, 0 Try.