Keep Cloud Native Moving: Building Japan's Platform for Open...
summary
KubeCon Europe draws 13,500 attendees from 3,500 organizations across 100 countries and CNCF reports about 20 million cloud native developers globally with 950,000 developers in Japan. Three projects graduate this year: Kyverno, OpenTelemetry, and Kubeflow. AI shifts compute to 2/3 inference now, with McKinsey predicting 93 gigawatts for inference by 2030. Concrete tooling mentioned includes vLLM and LLMD for horizontally scaled inference, an AI conformance program, and a $4 million GPU contribution to optimize GPU support.
AI-written summary · about
chapters (16)
- 0:00Welcome
- 1:05CNCF growth and updates
- 3:11Nvidia and new members
- 4:37Cloud native in Japan
- 6:23AI and cloud native
- 8:12KubeCon Japan winner
- 9:17Scaling AI inference
- 12:27LLMD optimization
- 14:11Model training trends
- 15:36Securing AI concerns
- 17:42Open source AI foundation
- 18:26AI conformance platform
- 20:09Get involved in CNCF
- 21:55CNCF ambassador programs
- 23:18Upcoming KubeCon events
- 24:46Closing
related talks
07mKeynote: Exploring the Frontiers of Cloud Native
15mKeynote: Fujitsu's Open Source Vision and Initiatives
07mSponsored Keynote: Cloud Native Best Practices
20mKeynote: What's Next? Getting Excited about Kubernetes in 2018
23mThe Makers of Marvels: How Developers Are Rebuilding the Enterprise, One Brick at a Time
17mKeynote: Welcome + Opening Remarks
16mKeynote: Accelerating Innovation Through Open Source
05mKeynote: Intuit’s AI-Native Platform: Scaling to 8x Velocity for 8,000 Developers