cs.LGAug 18, 2026

Efficient Resource Optimization for Split Federated Learning

Authors: Wei Wei, Xianhao Chen

Organizations: Department of Electrical and Computer Engineering, the University of Hong Kong, Pok Fu Lam, Hong Kong SAR, 999077

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Split federated learning (SFL) has emerged as a powerful paradigm for model training at the edge. However, SFL inherently involves discrete decision variables for model splitting and resource allocation, resulting in a challenging mixed-integer problem. Consequently, prior optimization schemes for SFL are either heuristic or computationally inefficient, which cannot handle large-scale user populations. To address this limitation, this work establishes an efficient optimization framework for SFL under resource-constrained networks. Our framework jointly optimizes model splitting and resource allocation to minimize training cost, which is defined as the weighted sum of latency and energy costs. We first study the model splitting problem and develop a polynomial-time algorithm that achieves the global optimum. Then, we extend the approach to the joint model splitting and resource allocation problem. In this case, we formulate it as a two-dimensional master problem and develop an efficient approximation method with a (1+ε)(1+ε)-approximation guarantee. Extensive experiments show that the proposed approach provides efficient solutions to strike the optimal energy--latency tradeoff.

Figures & tables

Fig. 1: Illustration of SFL workflow. (a) SFL scenario includes uploading, downloading and aggregation processes. (b) Since the server starts only after all clients finish local FP and activation uploading, and the next round begins only after all clients complete BP and gradient downloading, moderately reducing the GPU frequency and transmission power of faster clients does not affect the overall training time but can save energy.
Fig. 2: Impact of cut layer on energy and latency. The experiments are conducted with ResNet-50 on the CIFAR-10 dataset under non-IID settings. Other configurations are identical to experimental settings in Section VII .
NotationDescription
N\mathcal{N}The set of all clients
NNThe number of clients
LLThe number of layers in the global model
Lc,iL_{c,i}The cut layer of client ii
nin_{i}The number of GPU FLOPs per cycle of client ii
nSn_{\mathrm{S}}The number of GPU FLOPs per cycle of the edge server
TABLE I: Summary of Important Notations.
Fig. 3: Evaluating training performance on CIFAR-10 and MNIST using ResNet-50 and VGG-16 ( λ=100\lambda=100 ).
ParameterValueParameterValue
CC2020N0N_{0}3.98×10−213.98\times 10^{-21} W/Hz\mathrm{W/Hz}
JJ55λ\lambda100100
nin_{i}[0.2,3]×104[0.2,3]\times 10^{4} FLOPs/cycle\mathrm{FLOPs/cycle}nSn_{\mathrm{S}}3×1043\times 10^{4} FLOPs/cycle\mathrm{FLOPs/cycle}
GiG_{i}1×10−261\times 10^{-26} W/(cycle/s)3\mathrm{W/(cycle/s)^{3}}GSG_{\mathrm{S}}2×10−262\times 10^{-26} W/(cycle/s)3\mathrm{W/(cycle/s)^{3}}
fimax⁡f_{i}^{\max}625625 MHz\mathrm{MHz}fimin⁡f_{i}^{\min}5050 MHz\mathrm{MHz}
fSmax⁡f_{\mathrm{S}}^{\max}3.1353.135 GHz\mathrm{GHz}fSmin⁡f_{\mathrm{S}}^{\min}210210 MHz\mathrm{MHz}
TABLE II: Simulation Parameters
Fig. 4: The impact of uplink bandwidth and downlink bandwidth on cost function.
Fig. 5: The impact of GPU frequency on cost function.
Fig. 6: The impact of network resource fluctuation on cost function.
Fig. 7: The impact of number of clients on cost function and running time.
Fig. 8: Performance comparison under λ=0\lambda=0 and λ=∞\lambda=\infty .
More in this topic

Split Federated Learning

Split federated learning divides models between clients and servers to train collaboratively while keeping raw data local.

See all 37 papers

Explore similar work

CardsList
Feb 19, 2026

ASFL: An Adaptive Model Splitting and Resource Allocation Framework for Split Federated Learning

Federated learning (FL) enables multiple clients to collaboratively train a machine learning model without sharing their raw data. However, the limited computation resources of the clients may result in a high delay and energy consumption on training. In this paper, we propose an adaptive split federated learning (ASFL) framework over wireless networks. ASFL exploits the computation resources of the central server to train part of the model and enables adaptive model splitting as well as resource allocation during training. To optimize the learning performance (i.e., convergence rate) and efficiency (i.e., delay and energy consumption) of ASFL, we theoretically analyze the convergence rate and formulate a joint learning performance and resource allocation optimization problem. Solving this problem is challenging due to the long-term delay and energy consumption constraints as well as the coupling of the model splitting and resource allocation decisions. We propose an online optimization enhanced block coordinate descent (OOE-BCD) algorithm to solve the problem iteratively. Experimental results show that when compared with five baseline schemes, our proposed ASFL framework converges faster and reduces the total delay and energy consumption by up to 75% and 80%, respectively.

Chuiyang Meng, Ming Tang, Vincent W. S. Wong

Department of Electrical and Computer Engineering, The University of British Columbia, Vancouver, Canada · Department of Computer Science and Engineering, Southern University of Science and Technology, China

Split Federated LearningEnergy-Efficient MLWireless Resource Allocation
AskPDF
Mar 9, 2026

Split Federated Learning Architectures for High-Accuracy and Low-Delay Model Training

Can we find a network architecture for ML model training so as to optimize training loss (and thus, accuracy) in Split Federated Learning (SFL)? And can this architecture also reduce training delay and communication overhead? While accuracy is not influenced by how we split the model in ordinary, state-of-the-art SFL, in this work we answer the questions above in the affirmative. Recent Hierarchical SFL (HSFL) architectures adopt a three-tier training structure consisting of clients, (local) aggregators, and a central server. In this architecture, the model is partitioned at two partitioning layers into three sub-models, which are executed across the three tiers. Despite their merits, HSFL architectures overlook the impact of the partitioning layers and client-to-aggregator assignments on accuracy, delay, and overhead. This work explicitly captures the impact of the partitioning layers and client-to-aggregator assignments on accuracy, delay and overhead by formulating a joint optimization problem. We prove that the problem is NP-hard and propose the first accuracy-aware heuristic algorithm that explicitly accounts for model accuracy, while remaining delay-efficient. Simulation results on public datasets show that our approach can improve accuracy by 3%, while reducing delay by 20% and overhead by 50%, compared to state-of-the-art SFL and HSFL schemes.

Yiannis Papageorgiou, Yannis Thomas, Ramin Khalili +1

Department of Informatics, Athens University of Economics and Business, Athens, Greece · Huawei Technologies Duesseldorf Gmbh, Munich, Germany

Split Federated LearningFederated LearningDistributed TrainingHierarchical Federated LearningCommunication-Efficient Distributed Training
AskPDF
Oct 19, 2026

CoASFL: Coupled Consistency Optimization for Asynchronous Split Federated Learning

Split Federated Learning (SFL) has emerged as a promising paradigm for training large-scale deep models on resource-constrained edge devices. Although asynchronous SFL alleviates the straggler bottleneck caused by system heterogeneity, it inevitably introduces two coupled dimensions of inconsistency: horizontal inconsistency across clients and vertical inconsistency between clients and the server. These inconsistencies, further exacerbated by statistical heterogeneity, result in biased activation representations, model divergence, and severe client drift. To address these challenges, we propose CoASFL, an inconsistency-aware asynchronous SFL framework that incorporates a dynamic activation merging mechanism and a consistency-aware model aggregation mechanism. Specifically, CoASFL constructs more consistent activation sequences from heterogeneous clients to alleviate horizontal inconsistency, and dynamically adjusts aggregation weights according to client participation frequencies and progress gaps to mitigate vertical inconsistency. By jointly coordinating activation-level and model-level consistency, CoASFL effectively suppresses inconsistency propagation during asynchronous training. Extensive experiments on an 80-device heterogeneous edge testbed demonstrate that CoASFL consistently outperforms state-of-the-art SFL approaches, accelerating training by 1.61×\times to 3.74×\times and improving final model accuracy by up to 3.26% under heterogeneous scenarios.

Junhao Cheng, Yunming Liao, Hongli Xu +2

School of Computer Science and Technology, University of Science and Technology of China, Hefei, Anhui, China, 230027 · Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, Jiangsu, China, 215123

Split Federated LearningFederated LearningAsynchronous Federated LearningFederated Learning AggregationHeterogeneous Federated Learning
AskPDF