AI Trends

Trending story

AWS Launches Workload Recovery and Infrastructure Checks for SageMaker HyperPod

1 reports1 story updatesUpdated Nov 19 07:03

Understand this story

Story summary

The English summary is being updated. See the reporting timeline below.

Latest development · Report

Amazon SageMaker HyperPod Introduces Workload Recovery and Expanded Infrastructure Checks

Developments

1 confirmed developments

  1. English translation is being prepared.

    Source:AWS Machine Learning

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 19

  1. AWS Machine LearningFirst-party

    Amazon SageMaker HyperPod Introduces Workload Recovery and Expanded Infrastructure Checks

    Amazon SageMaker HyperPod has introduced workload recovery and infrastructure checks designed to detect distributed training failures earlier and increase the proportion of time spent on productive training. The new capabilities include MultiNodeTrainingValidation for EKS clusters, the Kubernetes Resource Monitoring Agent, and hung job detection for Slurm clusters. Hung job detection can log CloudWatch events, cancel jobs according to predefined rules, and then hand them back to Slurm for requeuing.

Related stories

  1. Related storiesAWS adds managed NVIDIA Dynamo support to SageMaker HyperPod Inference
  2. Related storiesAmazon SageMaker AI real-time endpoints support KV cache-aware routing
  3. Related storiesAWS adds AI data preparation capabilities to Amazon SageMaker AI
  4. Related storiesNVIDIA Vera Rubin NVL72 Makes Its Debut in MLPerf Training v6.1