The paper proposes two hybrid algorithms, FL+FSDP and FL+HSDP, that combine sharded data parallelism with federated learning-style aggregations to reduce communication overhead in large-scale AI training. By forming loosely‑coupled federation groups, the methods keep inter‑group traffic minimal while maintaining a bounded global batch size. Experiments on a Llama3.1 8B model trained on 512 A100 GPUs show up to 8.04× faster data processing and 4.48 lower evaluation perplexity compared to traditional sharded DP approaches.
The paper introduces FastFederatedLearning (FFL), an open‑source federated learning framework written in C/C++ that prioritizes performance and customizability. FFL allows users to define arbitrary communication graphs between clients and servers, and it consistently outperforms Intel OpenFL with speedups of 2.5x to 3.69x across x86‑64, ARM‑v8, and RISC‑V platforms. Future plans include a Python wrapper and middleware for diverse communication backends, aiming to support dynamic, evolving federations as long‑term services.
By Gianluca Mittone, Robert Birke, Marco Aldinucci
arXiv:2608. 14654v1 Announce Type: cross Abstract: Federated Learning (FL) is a collaborative paradigm that enables multiple devices to train a global model while preserving local data privacy.
By Hai Anh Tran, Cuong Ta, Truong X. Tran
arXiv:2608. 09208v1 Announce Type: cross Abstract: Decentralized intelligence systems with heterogeneous devices and limited coordination increasingly rely on decentralized federated learning (DFL).
By Van Truong Vo, Khoa Nguyen, Taehong Kim
arXiv:2411. 02908v2 Announce Type: replace Abstract: Scaling large language models (LLMs) demands extensive data and computing resources, which are traditionally constrained to data centers by the high-bandwidth requirements of distributed training.
By Lorenzo Sani, Alex Iacob, Zeyu Cao, Royson Lee, Bill Marino, Yan Gao, Dongqi Cai, Zexi Li, Wanru Zhao, Xinchi Qiu, Nicholas D. Lane
arXiv:2606. 19025v1 Announce Type: cross Abstract: Pre-training Large Language Models (LLMs) typically demands large-scale infrastructure with tightly coupled hardware accelerators.
By Lorenzo Sani, Zeyu Cao, Meghdad Kurmanji, Alex Iacob, Andrej Jovanovic, Yan Gao, Wanru Zhao, Nicholas D. Lane
Decentralized intelligence systems with heterogeneous devices and limited coordination increasingly rely on decentralized federated learning (DFL). However, DFL suffers from convergence inefficiency under data heterogeneity due to the use of a uniform learning rate (LR) that ignores layer-specific optimization needs.
arXiv:2606. 11169v1 Announce Type: cross Abstract: Large-scale model training increasingly relies on composing multiple parallelism strategies, such as data, pipeline, and expert parallelism, together with memory-saving optimizations like ZeRO.
By Megan Frisella, Shubham Tiwari, Andy Ruan, Yi Pan, Parker Gustafson, Mat Jacob, Gilbert Bernstein, Stephanie Wang
arXiv:2606. 06687v1 Announce Type: new Abstract: We investigate cluster formation, involving the number and composition of clusters, in decentralized federated learning (FL) with heterogeneous machine learning (ML) optimizers.
By Su Wang, Mung Chiang, H. Vincent Poor
arXiv:2512. 10236v2 Announce Type: replace-cross Abstract: Modern ML workloads demand distributing training and inference across multiple GPUs.
By Shagnik Pal, Shaizeen Aga, Suchita Pati, Mahzabeen Islam, Lizy K. John
FlexP-SFT introduces an aggregation-free framework for personalized split federated fine-tuning of large language models, eliminating the client-side aggregation step that traditionally causes communication bottlenecks and straggler issues. The method employs a layer‑flexible alignment strategy to balance personalization and generalization without global synchronization, and formulates split‑ratio selection as a resource‑aware discrete optimization problem. Experiments demonstrate that FlexP-SFT improves both accuracy and latency compared to baselines, achieving a superior resource‑accuracy trade‑off.
By Jiaxiang Geng, Tianjun Yuan, Pengchao Han, Ying Gao, Xianhao Chen, Bing Luo
ShardMeter is a lightweight analytical performance model that predicts end-to-end runtime for transformer-based workloads across sharded, distributed, and decentralized training setups. By taking a model’s characteristics and a target hardware topology as input, it estimates per-GPU and per-island throughput, training cost, total wall-clock time, and pinpoints performance bottlenecks. The model reveals diminishing-return regimes with increasing island size, quantifies compute- versus communication-bound scaling, evaluates hyperparameter trade-offs, and models cost-throughput for large-scale decentralized training, enabling rapid exploration of configuration space and near-optimal deployment plans.
By Tim Beringer (Technical University of Darmstadt), Patrick Diem (Technical University of Darmstadt), Felix Wolf (Technical University of Darmstadt), Arya Mazaheri (Technical University of Darmstadt, PanocularAI)