Protea: Client Profiling within Federated Systems using Flower

TL;DR

Protea is a client profiling tool for federated learning systems that improves simulation scalability and resource efficiency, enabling large-scale experiments with heterogeneous clients.

Contribution

We introduce Protea, a lightweight client profiling component within Flower that automatically collects system statistics and estimates resources for scalable federated learning simulations.

Findings

Achieves 1.66× faster wall-clock time in simulations

Improves GPU utilization by 2.6×

Enables large-scale experiments with heterogeneous clients

Abstract

Federated Learning (FL) has emerged as a prospective solution that facilitates the training of a high-performing centralised model without compromising the privacy of users. While successful, research is currently limited by the possibility of establishing a realistic large-scale FL system at the early stages of experimentation. Simulation can help accelerate this process. To facilitate efficient scalable FL simulation of heterogeneous clients, we design and implement Protea, a flexible and lightweight client profiling component within federated systems using the FL framework Flower. It allows automatically collecting system-level statistics and estimating the resources needed for each client, thus running the simulation in a resource-aware fashion. The results show that our design successfully increases parallelism for 1.66 $\times$ faster wall-clock time and 2.6$\times$ better GPU utilisation, which enables large-scale experiments on heterogeneous clients.