2025
Felipe Machado Schwanck, Marcos Tomazzoli Leipnitz, Joel Luís Carbonera, Juliano Araujo Wickboldt
A Framework for testing Federated Learning algorithms using an edge-like environment Journal Article
In: Future Generation Computer Systems, 166 , pp. 107626, 2025, ISSN: 0167-739X.
Abstract Links BibTeX Tags: Cloud Computing Federated Learning Fog & Edge Computing Machine Learning
@article{journal/fgcs/Schwanck25,
title = {A Framework for testing Federated Learning algorithms using an edge-like environment},
author = {Felipe Machado Schwanck and Marcos Tomazzoli Leipnitz and Joel Luís Carbonera and Juliano Araujo Wickboldt},
url = {https://www.sciencedirect.com/science/article/pii/S0167739X24005909},
doi = {10.1016/j.future.2024.107626},
issn = {0167-739X},
year = {2025},
date = {2025-05-01},
urldate = {2025-05-01},
journal = {Future Generation Computer Systems},
volume = {166},
pages = {107626},
abstract = {Federated Learning (FL) is a machine learning paradigm in which many clients cooperatively train a single centralized model while keeping their data private and decentralized. FL is commonly used in edge computing, which involves placing computer workloads (both hardware and software) as close as possible to the edge, where data are created and where actions are occurring, enabling faster response times, greater data privacy, and reduced data transfer costs. However, due to the heterogeneous data distributions/contents of clients, it is non-trivial to accurately evaluate the contributions of local models in global centralized model aggregation. This is an example of a major challenge in FL, commonly known as data imbalance or class imbalance. In general, testing and evaluating FL algorithms can be a very difficult and complex task due to the distributed nature of the systems. In this work, a framework is proposed and implemented to evaluate FL algorithms in a more easy and scalable way. This framework is evaluated over a distributed edge-like environment managed by a container orchestration platform (i.e. Kubernetes).},
keywords = {Cloud Computing, Federated Learning, Fog & Edge Computing, Machine Learning},
pubstate = {published},
tppubtype = {article}
}
Federated Learning (FL) is a machine learning paradigm in which many clients cooperatively train a single centralized model while keeping their data private and decentralized. FL is commonly used in edge computing, which involves placing computer workloads (both hardware and software) as close as possible to the edge, where data are created and where actions are occurring, enabling faster response times, greater data privacy, and reduced data transfer costs. However, due to the heterogeneous data distributions/contents of clients, it is non-trivial to accurately evaluate the contributions of local models in global centralized model aggregation. This is an example of a major challenge in FL, commonly known as data imbalance or class imbalance. In general, testing and evaluating FL algorithms can be a very difficult and complex task due to the distributed nature of the systems. In this work, a framework is proposed and implemented to evaluate FL algorithms in a more easy and scalable way. This framework is evaluated over a distributed edge-like environment managed by a container orchestration platform (i.e. Kubernetes).
2022
Fernando Ferreira Remde, Juliano Araujo Wickboldt
Analyzing Federated Learning Performance in Distributed Edge Scenarios Inproceedings
In: 27º Workshop de Gerência e Operação de Redes e Serviços, WGRS 2022, Virtual Conference, May 23-27, 2022, pp. 155-168, SBC, 2022, ISSN: 2595-2722.
Abstract Links BibTeX Tags: Federated Learning Fog & Edge Computing Machine Learning
@inproceedings{conf/wgrs/Remde22,
title = {Analyzing Federated Learning Performance in Distributed Edge Scenarios},
author = {Fernando Ferreira Remde and Juliano Araujo Wickboldt},
url = {https://www.inf.ufrgs.br/~jwickboldt/wp-content/uploads/Analyzing-Federated-Learning-Performance-in-Distributed-Edge-Scenarios-WGRS-2022-Camera-Ready.pdf
https://sol.sbc.org.br/index.php/wgrs/article/view/21484},
doi = {10.5753/wgrs.2022.223574},
issn = {2595-2722},
year = {2022},
date = {2022-05-23},
urldate = {2022-05-23},
booktitle = {27º Workshop de Gerência e Operação de Redes e Serviços, WGRS 2022, Virtual Conference, May 23-27, 2022},
pages = {155-168},
publisher = {SBC},
abstract = {Federated Learning is a machine learning paradigm where many clients cooperatively train a single centralized model while keeping their data private and decentralized. This novel paradigm imposes many challenges, such as dealing with data that is not independent and identically distributed, spread among multiple clients that are not synchronized and may have limited computing power. These clients are often edge devices such as smartphones and sensors, which form a system that is heterogeneous, highly distributed by nature and difficult to manage. This work proposes an architecture for running federated learning experiments in a distributed edge-like environment. Based on this architecture, a set of experiments are conducted to analyze how the overall system performance is affected by different configuration parameters and varied number of connected clients.},
keywords = {Federated Learning, Fog & Edge Computing, Machine Learning},
pubstate = {published},
tppubtype = {inproceedings}
}
Federated Learning is a machine learning paradigm where many clients cooperatively train a single centralized model while keeping their data private and decentralized. This novel paradigm imposes many challenges, such as dealing with data that is not independent and identically distributed, spread among multiple clients that are not synchronized and may have limited computing power. These clients are often edge devices such as smartphones and sensors, which form a system that is heterogeneous, highly distributed by nature and difficult to manage. This work proposes an architecture for running federated learning experiments in a distributed edge-like environment. Based on this architecture, a set of experiments are conducted to analyze how the overall system performance is affected by different configuration parameters and varied number of connected clients.