Amazon SageMaker HyperPod now supports AMI versioning and auto-patching

Amazon SageMaker HyperPod now supports AMI versioning and auto-patching to improve cluster security and consistency while reducing operational burden.

• View exact AMI versions across all instance groups and nodes in semantic versioning format
• Detect version drift quickly and roll back to previous versions using UpdateClusterSoftware API
• Auto-patching applies backward-compatible security patches when nodes are idle without disrupting workloads
• Critical packages like NVIDIA driver and CUDA never upgrade to different major or minor versions
• New AMI support policy publishes timelines for security patch availability
• Available for EKS-orchestrated HyperPod clusters in all supported AWS Regions

These capabilities help cluster administrators maintain security and consistency while eliminating manual patching overhead for long-running AI/ML training jobs.