Amazon SageMaker AI now supports instance preference lists for training and processing jobs, allowing users to specify prioritized instance types and automatically run jobs on the first available configuration.

<div><p>This article announces instance preference lists for Amazon SageMaker AI training and processing jobs, enabling faster compute capacity allocation.</p><ul><li>Specify prioritized lists of instance types and sizes for job submission</li><li>SageMaker automatically launches jobs on the first available configuration</li><li>Eliminates need for complex retry logic or concurrent job submissions</li><li>Configure capacity sourcing from on-demand or SageMaker Flexible Training Plans</li><li>Available in all AWS Regions where SageMaker is available</li></ul><p>Instance preference lists reduce wait times and manual retrying for high-demand GPU workloads during peak periods.</p></div>


Amazon SageMaker AI now supports instance preference lists for training and processing jobs

Related articles