Go with the Flow

Churn-Tolerant Decentralized Training of Large Language Models

Conference Paper (2025)
Author(s)

Nikolay Blagoev (University of Neuchâtel, Worker Thread)

Bart Cox (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Jérémie Decouchant (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Lydia Y. Chen (University of Neuchâtel)

Research Group
Data-Intensive Systems
DOI related publication
https://doi.org/10.1109/BigData66926.2025.11402098 Final published version
More Info
expand_more
Publication Year
2025
Language
English
Research Group
Data-Intensive Systems
Pages (from-to)
3365-3375
Publisher
IEEE
ISBN (electronic)
9798331594473
Event
2025 IEEE International Conference on Big Data, BigData 2025 (2025-12-08 - 2025-12-11), Macau, China
Downloads counter
37
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Motivated by the emergence of large language models (LLMs) and the importance of democratizing their training, we propose GWTF, the first fully decentralized churn-tolerant practical training framework for LLMs. Differently from existing distributed and federated training frameworks, GWTF enables the efficient collaborative training of a LLM on heterogeneous clients that volunteer their resources. In addition, GWTF addresses node churn, i.e., clients joining or leaving the system at any time, and network instabilities, i.e., network links becoming unstable or unreliable. The core of GWTF is a novel decentralized flow algorithm that finds the most effective routing that maximizes the number of microbatches trained with the lowest possible delay. We extensively evaluate GWTF on GPT-like and LLaMalike models and compare it against the prior art. Our results indicate that GWTF reduces the training time by up to 45 % in realistic and challenging scenarios that involve heterogeneous client nodes distributed over 10 different geographic locations with a high node churn rate.

Files

Go_With_the_Flow_Churn-Toleran... (pdf)
(pdf | 1.29 Mb)
- Embargo expired in 06-09-2026
– Personal use only – Dutch Copyright Act (Article 25fa)