Project Lightning Talk: Empowering Federated Learning with Multi-Cluster Management for Privacy and Efficiency
Lightningat KubeCon + CloudNativeCon Europe 2025 · on Tuesday, 2025-04-01· at 12:20· Platinum Suite | Level 3·
summary
Shows an end-to-end integration of Open-Cluster-Management with Federated Learning that moves model updates not data, cutting cross-cluster bandwidth and storage needs while keeping private data local. Uses OCM hub as FL aggregator and managed clusters as clients, supports Flower OpenFL and FATE via a controller with CRD to manage workflows, requires containerized training images, and surfaces statuses waiting→running→complete. The built controller schedules server and clients across hub‑spoke clusters and demonstrates successful converged global models without data transfer.
AI-written summary · about
related talks
42mFederated Learning: Unlocking the Values of Discrete Data for Enterprise
48mFederated ML for Edge Applications
32mInnovation in the Age of Regulation: Federated Learning with Flower
31mFrom Theory to Practice: Advances in Real-World Federated Learning
21mWhat, Why and How of Federated Machine Learning – Implementation Using...
31mFederated Machine Learning with Kubernetes
32mOpenFL: A Federated Learning Project to Power (and Secure) Your Projects
28mA Privacy Enhancing Federated Learning Framework for Multi-Party Distributed ..