FedBed: Benchmarking Federated Learning over Virtualized Edge Testbeds

Moysis Symeonides, Fotis Nikolaidis, Demetris Trihinas, George Pallis, Marios D. Dikaiakos, Angelos Bilas

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    Abstract

    Federated Learning has become the de facto paradigm for training AI models under a distributed modality where the computational effort is spread across several clients without sharing local data. Despite its distributed nature, enabling FL in an Edge-Cloud continuum is challenging with resource and network heterogeneity, different AI models and libraries, and non-uniform data distributions, all hampering QoS and limiting innovation potential. This work introduces FedBed, a testing framework that enables the rapid and reproducible benchmarking of FL deployments on virtualized testbeds. FedBed aids users in assessing the numerous trade-offs that result from combining a variety of FL software and infrastructure configurations in Edge-Cloud settings. This reduces the time-consuming process that includes the setup of either a virtual physical or emulation testbed, experiment configurations, and the monitoring of the resulting FL testbed.

    Original languageEnglish
    Title of host publication16th IEEE/ACM International Conference on Utility and Cloud Computing, UCC 2023
    PublisherAssociation for Computing Machinery, Inc
    ISBN (Electronic)9798400702341
    DOIs
    Publication statusPublished - 4 Dec 2023
    Event16th IEEE/ACM International Conference on Utility and Cloud Computing, UCC 2023 - Taormina, Italy
    Duration: 4 Dec 20237 Dec 2023

    Publication series

    Name16th IEEE/ACM International Conference on Utility and Cloud Computing, UCC 2023

    Conference

    Conference16th IEEE/ACM International Conference on Utility and Cloud Computing, UCC 2023
    Country/TerritoryItaly
    CityTaormina
    Period4/12/237/12/23

    Keywords

    • edge computing
    • federated learning

    Fingerprint

    Dive into the research topics of 'FedBed: Benchmarking Federated Learning over Virtualized Edge Testbeds'. Together they form a unique fingerprint.

    Cite this