Throughput-Optimal Topology Design for Cross-Silo Federated Learning

arXiv2010.12229MaRDI QIDQ6352057

Author name not available (Why is that?)

Publication date: 23 October 2020

Abstract: Federated learning usually employs a client-server architecture where an orchestrator iteratively aggregates model updates from remote clients and pushes them back a refined model. This approach may be inefficient in cross-silo settings, as close-by data silos with high-speed access links may exchange information faster than with the orchestrator, and the orchestrator may become a communication bottleneck. In this paper we define the problem of topology design for cross-silo federated learning using the theory of max-plus linear systems to compute the system throughput---number of communication rounds per time unit. We also propose practical algorithms that, under the knowledge of measurable network characteristics, find a topology with the largest throughput or with provable throughput guarantees. In realistic Internet networks with 10 Gbps access links for silos, our algorithms speed up training by a factor 9 and 1.5 in comparison to the master-slave architecture and to state-of-the-art MATCHA, respectively. Speedups are even larger with slower access links.

Has companion code repository: https://github.com/omarfoq/communication-in-cross-silo-fl

This page was built for publication: Throughput-Optimal Topology Design for Cross-Silo Federated Learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6352057)