Keynote: Enabling AI with Cloud Native
summary
CNCF frames concrete work to run GenAI at scale: hardware slicing for GPUs/TPUs/FPGAs via dynamic resource allocation, improved NUMA alignment via KEP-3545, and workflow APIs that land in Kubernetes (batch jobs in 1.24, 100,000 concurrent pods in 1.26) driving 100,000 concurrent pods and Kubeflow 1.8 adoption. Projects to watch: Volcano, batch working groups, Carada (multicluster), KubeEdge/Sedna for edge inference and multicluster orchestration.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:28Jorge Castro Introduction
- 2:24Next Trillion Core Hours
- 3:37What does Open look like
- 4:02Kubernetes mindset
- 4:44Hardware enablement
- 6:46Application development
- 7:33Integrations
- 8:05New Primitives
- 8:25Kubernetes is already the web server
- 8:47scrambling up the stack
- 9:11multiple organizations
- 9:38building a scaffolding
related talks
10mKeynote: Enabling AI with Cloud Native
07mKeynote: Exploring the Frontiers of Cloud Native
09mKeynote: Kubernetes for GenAI: The New Operating System of the AI-First World
15mKeynote: From Platforms to AI Factories. Has Kubernetes Solved It?
14mKeynote: Building the Infrastructure that Powers the Future of AI
06mSponsored Keynote: Building Your AI Factory: Wh...
35mPanel: Beyond the Clouds: Charting the Course for AI in the CloudNative World
20mOperating Frontier Intelligence at Scale