The Different Shades of using KubeRay with Kubernetes

Описание: With KubeRay, a single command can get your ray-cluster up and running on Kubernetes. In this talk we start by demo-ing to the users how to deploy KubeRay and create their first cluster in seconds. And then we go beyond the basic use case of using Kubectl and Helm to create ray-clusters. We present the extended API to manage ray-clusters from an end-user perspective. We also present the clients libraries allowing your application running inside a ray-cluster to scale its own ray-cluster on demand, or even spin up new ray-clusters and manage them programmatically. This will include the newly added Python-Client KubeRay library (https://github.com/ray-project/kubera...) which developers will appreciate the ease with which they can setup clusters (in K8s) without dealing with the K8s platform or Kubectl and Yaml files.

At the end of the talk, you will be familiar with how ray clustering works on Kubernetes and all the different mechanisms the KubeRay project offers you to create and manage ray-clusters.

Find the slide deck here: https://drive.google.com/file/d/1963X...

About Anyscale
---
Anyscale is the AI Application Platform for developing, running, and scaling AI.

https://www.anyscale.com/

If you're interested in a managed Ray service, check out:
https://www.anyscale.com/signup/

About Ray
---
Ray is the most popular open source framework for scaling and productionizing AI workloads. From Generative AI and LLMs to computer vision, Ray powers the world’s most ambitious AI workloads.
https://docs.ray.io/en/latest/


#llm #machinelearning #ray #deeplearning #distributedsystems #python #genai