Federated Learning in the Edge-Cloud Continuum: A Task-Based Approach with Colony

The edge-cloud continuum enables distributed machine learn- ing by leveraging the complementary strengths of edge devices and cen- tralized cloud resources. Federated Learning (FL) has emerged as a key paradigm in this context, allowing collaborative model training across multiple parties without sharing raw data, thus preserving privacy, re- ducing communication costs, and supporting compliance with data pro- tection regulations. However, orchestrating FL workflows across hetero- geneous edge-cloud environments introduces significant challenges re- lated to task coordination, resource management, and scalability. In this paper, we propose using the Colony framework to address these chal- lenges through a task-based approach to FL. Colony allows develop- ers to define an FL workflow as parallel tasks automatically scheduled across heterogeneous resources. We show how this task-based model sup- ports core FL operations—such as local training, model aggregation, and synchronization—within a unified execution framework. Experiments on a medical imaging use case demonstrate that Colony enables scalable and efficient orchestration of FL tasks across heterogeneous environments while ensuring that sensitive data remain local. These results highlight the applicability and advantages of task-based programming models for privacy-preserving machine learning across the compute continuum.

HeteroPar 2025, 23rd International Workshop