Home icon

Powering the Next Generation of AI Workloads on Amazon EKS with Anyscale

Containers Blog



This article discusses a new version of Ray called RayTurbo, which is an open-source framework for managing and executing AI workloads. The key highlights are:

Specifically, the article covers:

  • RayTurbo is 4.5x faster than the previous Ray version for data processing and offers optimizations to run online model serving with up to 50% fewer nodes.
  • Anyscale has integrated RayTurbo natively with Kubernetes, allowing customers to connect their Amazon EKS clusters to Anyscale's platform and manage RayTurbo across their EKS clusters.
  • The combination of Amazon EKS and RayTurbo provides customers with next-generation performance, reliability, and ease of use for training and running AI applications on AWS infrastructure, including AWS Trainium and AWS Inferentia chips.
  • Conclusion: RayTurbo on EKS is currently available to select customers, and Anyscale and AWS are working to bring this experience to more customers in the coming months.


Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

May 29
2025
Introducing AI on EKS: powering scalable AI workloads with Amazon EKS
Jul 16
2025
Amazon EKS enables ultra scale AI/ML workloads with support for 100K nodes per cluster
Sep 2
2025
Unlocking next-generation AI performance with Dynamic Resource Allocation on Amazon EKS and Amazon EC2 P6e-GB200
Dec 1
2025
Announcing Amazon EKS Capabilities for workload orchestration and cloud resource management

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.