FedFog: Resource-Aware Federated Learning in Edge and Fog Networks

TL;DR

FedFog is a simulation framework that enables resource-aware federated learning in edge and fog networks, improving model convergence, latency, and energy efficiency in IoT scenarios.

Contribution

It introduces FedFog, a novel simulation tool integrating federated learning with serverless edge-fog architectures, supporting adaptive scheduling and privacy-preserving data flow.

Findings

Accelerates model convergence in simulations

Reduces latency and energy consumption

Enhances scalability of federated learning in edge environments

Abstract

As edge and fog computing become central to modern distributed systems, there's growing interest in combining serverless architectures with privacy-preserving machine learning techniques like federated learning (FL). However, current simulation tools fail to capture this integration effectively. In this paper, we introduce FedFog, a simulation framework that extends the FogFaaS environment to support FL-aware serverless execution across edge-fog infrastructures. FedFog incorporates an adaptive FL scheduler, privacy-respecting data flow, and resource-aware orchestration to emulate realistic, dynamic conditions in IoT-driven scenarios. Through extensive simulations on benchmark datasets, we demonstrate that FedFog accelerates model convergence, reduces latency, and improves energy efficiency compared to conventional FL or FaaS setups-making it a valuable tool for researchers exploring scalable, intelligent edge systems.