Coastal lidar tiles already sit on Amazon EFS. Recopying that tree onto each SageMaker instance root volume would move several terabytes before training starts. How should the training resource read the tiles?
Select an answer to reveal the explanation.
Short Explanation
Think of coastal lidar already sitting on EFS, and a recopy onto each SageMaker root volume moving several terabytes first. Mount the existing EFS as the training data source. Instances open the shared tree. Recopying is the long way around.
Full Explanation
Amazon SageMaker can mount an existing Amazon EFS file system as the training job data source, so instances read the shared tree instead of recopying terabytes onto each root volume. Recopying wastes time and disk. Rekognition and Polly are not EFS mounts.