Scalable Machine Learning with AWS Batch

When: July 19, 2021
1:00 PM (PT) | 4:00 PM (ET) | 10:00 PM (CET)

Type: Service How To | Level: 300

Learn how you can accelerate machine learning training on AWS Batch with Amazon EC2 P4d instances and optimize costs with Amazon EC2 Spot Instances.

Join us as we walk through the methodology and architecture based on AWS Batch for training and inferencing machine learning models at scale.

The proposed approach will be demonstrated through an unsupervised learning problem on a high-dimensional dataset and will provide working code for the runs.

What you’ll learn

Learn how to divide ML workloads into atomic steps to leverage a combination of single & array jobs for a massive parallel execution

Understand resilient implementation with queues and dependencies to schedule the multiple iterations using batch queues

See how to leverage a mix of on-demand and spot instance compute environments to lower costs

Speaker(s): Venkatesh Rajagopalan, Principal Solutions Architect, AWS; Mahadevan Balasubramaniam, Principal Solutions Architect, AWS

Who Should Attend: 

BMachine Learning Developers, Data Scientists

WordPress Appliance - Powered by TurnKey Linux