Help improve this page
To contribute to this user guide, choose the Edit this page on GitHub link that is located in the right pane of every page.
Recipes to optimize your Amazon EKS cluster for AI/ML workloads
Register
This section is designed to provide bite-sized recipes for optimizing your Amazon EKS cluster, particularly for AI/ML workloads involving specialized hardware. Each recipe includes prerequisites, step-by-step procedures, and deployment considerations.
-
Target AI/ML workloads to specific nodes with labels — Target workloads to the right nodes using the labels that Amazon EKS applies automatically.
-
Place distributed AI/ML workloads with network topology labels — Place the Pods of a distributed job close to each other using network topology labels.
-
Prevent Pods from being scheduled on accelerated nodes — Keep workloads off accelerated nodes by adding taints, with EKS Auto Mode, Karpenter, and managed node groups.
-
Configure hugepages in Amazon EKS clusters — Configure hugepages for memory-intensive workloads.
-
Use AWS Inferentia instances with Amazon EKS for Machine Learning — Create an inference cluster with AWS Inferentia instances.