Jicheng Yuan, Duc Manh Nguyen - Federated Learning for IoT Edge devices
The hands-on workshop "Federated Learning for IoT Edge Devices" explores the application of federated learning in the context of Internet of Things (IoT). It focuses on a decentralized machine learning paradigm in a Federated Learning manner, allowing edge devices, such as Jetson Nano and Jetson Xavier to collaboratively train models without sharing raw data. This approach enhances privacy and efficiency by conducting model training locally on devices, minimizing the need for centralized data storage. The workshop highlights the potential of federated learning to empower IoT edge devices with intelligent decision-making capabilities while preserving data security and privacy.
Thomas Pusztai - FTUW SLO management framework
This workshop first introduced the concepts of the open-source Polaris SLO Framework. This orchestrator-independent framework is designed for creating high-level Service Level Objectives (SLOs) and enforcing them using either generic or application-specific elasticity strategies in cloud and edge environments. The hands-on part guided participants through the creation of a custom horizontal elasticity strategy and an average CPU usage SLO for scaling a demo application.
