Google Cloud and Anyscale are collaborating to make Apache Ray a key compute engine for AI and machine learning workloads in the cloud. The plan is to enable deeper integration between Ray and Google Kubernetes Engine (GKE), simplifying the deployment of distributed AI tasks at scale.
What Is Apache Ray?
Apache Ray is an open-source framework for distributed computing, designed primarily for Python developers. It allows for the scaling of AI applications by distributing tasks across clusters without requiring complex infrastructure code. Ray supports libraries for reinforcement learning, hyperparameter tuning, and data processing, making it a versatile tool for AI engineers and researchers.
Google Cloud and Anyscale’s Integration Plan
At Google Cloud Next 2024, the companies announced their intent to bring native support for Ray to GKE. This integration will eliminate the need for manual cluster setup for Ray applications. Instead, GKE will offer standardized support for deploying and managing Ray clusters efficiently.
This is a significant advancement for organizations running large AI projects in the cloud. With Ray on GKE, users can expect:
- Seamless scaling of AI/ML workloads
- Minimal infrastructure configuration
- Built-in support for GPU and TPU resources
- Enhanced multi-cluster resource management
Benefits for Developers
The planned integration aims to lower the barrier for developers launching large-scale AI workloads in the cloud. Traditionally, deploying Ray clusters required Kubernetes expertise and manual configuration. With GKE support, developers will be able to:
- Deploy and scale Ray clusters through Google Cloud Console or gcloud CLI
- Integrate with Vertex AI for model training, deployment, and monitoring
- Automatically utilize Google Cloud’s logging and monitoring tools
Anyscale’s Role and Support
Anyscale, founded by the creators of Ray, also announced expanded collaboration with Google Cloud. The company will offer enterprise-grade services for users deploying Ray on GKE, including enhanced security, support, and SLA-backed performance.
According to Anyscale, this partnership will help businesses move AI projects from prototype to production faster and more reliably.
Availability Timeline
This integration is currently under development. A public beta version is expected to be available in the second half of 2025. Google Cloud and Anyscale are inviting interested developers to sign up for early access and provide feedback.
This move signals a broader trend of making powerful AI infrastructure more accessible to cloud-native developers, helping to accelerate innovation in machine learning and AI deployment.
Experts in development, customization, release and production support of mobile and desktop applications and games. Offering a well-balanced blend of technology skills, domain knowledge, hands-on experience, effective methodology, and passion for IT.