Anyscale Inc., creator of the open-source distributed computing platform Ray, today announced a new partnership with Microsoft Corp. and the private preview of a new artificial intelligence-native compute service co-developed with Microsoft and delivered as a fully managed, first-party offering on Azure.

The new compute service, powered by Ray, offers an open-source distributed compute framework for AI that gives enterprises a streamlined, high-performance experience for building, running and scaling up advanced AI workloads.

Ray is an open-source distributed computing framework that allows developers to scale AI and Python workloads from a single laptop to large clusters. The service unifies data processing, model training and inference across central processing units and graphics processing units for building and deploying complex, multimodal AI applications.

The new offering with Microsoft seeks to address the issue where organizations face limitations from legacy computing systems not designed for today’s dynamic, large-scale AI workloads when embedding AI into customer interactions. Anyscale argues that modern AI requires high-performance computing that can scale dynamically across diverse data, models and accelerators, which are capabilities that traditional computing architectures struggle to deliver efficiently.

The new Anyscale-powered service on Azure addresses these challenges by bringing AI-native computing to every enterprise. It combines the performance and flexibility of Ray’s distributed compute engine, as well as Anyscale’s developer tooling and cluster management expertise, and the security, scale and trust of Azure.

With Anyscale running on Azure, it turns Ray’s AI-native computing framework into a fully managed, first-party service that streamlines setup and development. That, the companies say, removes the operational complexity of managing clusters and accelerates workloads with the Anyscale Runtime, all within Azure’s infrastructure.

Anyscale says its managed service on Azure provides AI teams with developer velocity, production resilience, cost-efficient processing and security and governments.

Development teams can use the service to quickly deploy clusters from Azure Portal, run interactive development with cluster-backed integrated developer environments and debug distributed applications with advanced dashboards. Fully managed, fault-tolerant Ray clusters are offered for both batch and low-latency serving directly inside Azure customers’ Azure Kubernetes Service.

Anyscale claims the service delivers up to 10 times faster performance compared wit self-managed Ray OSS without code changes. It also runs directly inside a customer’s Azure account to ensure that data, compute and AI models remain under their full control and comply with Azure security policies.

Azure customers can set up and manage Anyscale directly from the Azure Portal, run their Ray-powered AI workloads on Azure Kubernetes Service, and take advantage of unified billing through Azure.

“Enterprises want flexibility and control as they scale AI workloads,” said Brendan Burns, corporate vice president for cloud-native compute at Microsoft and co-creator of Kubernetes. “By bringing Ray to Azure Kubernetes Service through Anyscale on Azure, we’re giving developers the ability to use familiar cloud-native patterns to build and scale AI applications.”

Image: SiliconANGLE/Ideogram