Anyscale on Azure: Build, Train, and Serve AI in Your Own Tenant
summary
Anyscale on Azure is an Azure‑native, co‑engineered platform that runs Ray on AKS with the control plane in Azure and the runtime inside the customer’s AKS subscription, keeping data and models stay in your Azure tenant. It exposes observability and Entra/Azure AD permissions, supports mixed CPU/GPU pipelines (video decode on CPU plus LLMs on GPU) and can scale from 1 to 1,000 GPUs for distributed training; customers process petabytes of video and millions of kilometers of satellite imagery, showing the system works in production.
AI-written summary · about
chapters (8)
- 0:00Introduction to Anyscale on Azure
- 1:30Priya’s AI infrastructure challenge
- 3:36Anyscale on Azure architecture overview
- 5:43Industries and use cases adoption
- 7:26Partnership with Microsoft Azure
- 8:49Demo setup in Azure portal
- 12:19Security and enterprise deployment
- 13:20Call to action and closing
related talks
22mAI @Scale
21mHow Ray and Anyscale Make it Easy to do Massive-scale ML on Aerial Imagery
31mManaging Large-Scale Ray Deployments: Cloud, On-Prem, and Kubernetes
28mScaling AI & Machine Learning Workloads With Ray on AWS, Kubernetes, & BERT
48mGaming Industry use cases for Azure & AI
16mCracking the code: AI workloads at scale on AKS
20mThe Quick Journey to Using Ray: How We Implement Ray and Anyscale to Speed up our ML Processes
50mAzure AI, Power new possibilities for every organization