{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/132539"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/132539","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Energy and time efficient federated learning","abstract":"Over the past decade, the volume of data generated by edge devices has grown exponentially as the number of edge devices surges. Federated learning (FL) enables on-device training while preserving privacy, but edge devices typically operate under tight time and energy budgets, highlighting the need for time- and energy-efficient FL algorithms. While prior work focuses on either compute or communication costs, our hardware characterization shows that, although per-bit communication is far more expensive than per-bit computation, the relative importance of compute and communication costs depends on the communication period. Therefore, both costs must be considered before we find the optimal communication period. In this work, we formulate the optimal time and energy efficiency of FL systems into two separate optimization problems, characterized by four design variables: communication period C, GPU frequency f_GPU, number of clients K, and number of global rounds G. We solve both problems analytically and empirically using our simulation framework featuring a maximum of 10 Jetson clients with IID Cifar-10 dataset. Our empirical results show that time efficiency is optimal at the maximum available GPU frequency f^(max)_GPU, while energy efficiency is optimal at the highest GPU frequency with minimum supply voltage. Ablation studies reveal that optimal K for time efficiency increases with target accuracy; while optimal K for energy efficiency is always the smallest feasible K. The ablation also shows that optimal C for both time and energy efficiency increase with target accuracy. Based on the optimal hyperparameter setting, we find that optimal communication latency dominates only in low-accuracy regions (20%−29%); whereas optimal compute latency dominates in high-accuracy regions (30%−69%). On the other hand, optimal compute energy dominates across all accuracy regions. These results imply that compute costs are the primary concern for system efficiency under homogeneous FL settings.","abstract_html":"Over the past decade, the volume of data generated by edge devices has grown exponentially as the number of edge devices surges. Federated learning (FL) enables on-device training while preserving privacy, but edge devices typically operate under tight time and energy budgets, highlighting the need for time- and energy-efficient FL algorithms. While prior work focuses on either compute or communication costs, our hardware characterization shows that, although per-bit communication is far more expensive than per-bit computation, the relative importance of compute and communication costs depends on the communication period. Therefore, both costs must be considered before we find the optimal communication period. In this work, we formulate the optimal time and energy efficiency of FL systems into two separate optimization problems, characterized by four design variables: communication period C, GPU frequency f_GPU, number of clients K, and number of global rounds G. We solve both problems analytically and empirically using our simulation framework featuring a maximum of 10 Jetson clients with IID Cifar-10 dataset. Our empirical results show that time efficiency is optimal at the maximum available GPU frequency f^(max)_GPU, while energy efficiency is optimal at the highest GPU frequency with minimum supply voltage. Ablation studies reveal that optimal K for time efficiency increases with target accuracy; while optimal K for energy efficiency is always the smallest feasible K. The ablation also shows that optimal C for both time and energy efficiency increase with target accuracy. Based on the optimal hyperparameter setting, we find that optimal communication latency dominates only in low-accuracy regions (20%−29%); whereas optimal compute latency dominates in high-accuracy regions (30%−69%). On the other hand, optimal compute energy dominates across all accuracy regions. These results imply that compute costs are the primary concern for system efficiency under homogeneous FL settings.","abstract_has_math":false,"creators":["Hwang, Jing-Teng"],"institution":"University of Illinois Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Electrical & Computer Engr","degree_department":null,"school":null,"contributors":["Shanbhag, Naresh R."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-12","date_published":"2025-12","updated_at":"2026-07-22T22:25:07Z","subjects":["Edge Device","Federated Learning"],"languages":["en"],"rights":["Copyright 2025 Jing-Teng Hwang"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/132539","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Shanbhag, Naresh R."]},{"key":"dc:creator","label":"Author","values":["Hwang, Jing-Teng"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025-12","2025-12-02"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical & Computer Engr"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Edge Device","Federated Learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2025 Jing-Teng Hwang"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/132539"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Over the past decade, the volume of data generated by edge devices has grown exponentially as the number of edge devices surges. Federated learning (FL) enables on-device training while preserving privacy, but edge devices typically operate under tight time and energy budgets, highlighting the need for time- and energy-efficient FL algorithms. While prior work focuses on either compute or communication costs, our hardware characterization shows that, although per-bit communication is far more expensive than per-bit computation, the relative importance of compute and communication costs depends on the communication period. Therefore, both costs must be considered before we find the optimal communication period. In this work, we formulate the optimal time and energy efficiency of FL systems into two separate optimization problems, characterized by four design variables: communication period C, GPU frequency f_GPU, number of clients K, and number of global rounds G. We solve both problems analytically and empirically using our simulation framework featuring a maximum of 10 Jetson clients with IID Cifar-10 dataset. Our empirical results show that time efficiency is optimal at the maximum available GPU frequency f^(max)_GPU, while energy efficiency is optimal at the highest GPU frequency with minimum supply voltage. Ablation studies reveal that optimal K for time efficiency increases with target accuracy; while optimal K for energy efficiency is always the smallest feasible K. The ablation also shows that optimal C for both time and energy efficiency increase with target accuracy. Based on the optimal hyperparameter setting, we find that optimal communication latency dominates only in low-accuracy regions (20%−29%); whereas optimal compute latency dominates in high-accuracy regions (30%−69%). On the other hand, optimal compute energy dominates across all accuracy regions. These results imply that compute costs are the primary concern for system efficiency under homogeneous FL settings.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2026-02-19 without embargo terms","The student, Jing-Teng Hwang, accepted the attached license on 2025-11-26 at 22:29.","The student, Jing-Teng Hwang, submitted this Thesis for approval on 2025-11-26 at 23:00.","This Thesis was approved for publication on 2025-12-02 at 10:13.","DSpace SAF Submission Ingestion Package generated from Vireo submission #22975 on 2026-02-19 at 18:25:32"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Energy and time efficient federated learning"]}]}],"canonical_facts":{"dc:contributor":["Shanbhag, Naresh R."],"dc:creator":["Hwang, Jing-Teng"],"dc:date":["2025-12","2025-12-02"],"dc:description":["Over the past decade, the volume of data generated by edge devices has grown exponentially as the number of edge devices surges. Federated learning (FL) enables on-device training while preserving privacy, but edge devices typically operate under tight time and energy budgets, highlighting the need for time- and energy-efficient FL algorithms. While prior work focuses on either compute or communication costs, our hardware characterization shows that, although per-bit communication is far more expensive than per-bit computation, the relative importance of compute and communication costs depends on the communication period. Therefore, both costs must be considered before we find the optimal communication period. In this work, we formulate the optimal time and energy efficiency of FL systems into two separate optimization problems, characterized by four design variables: communication period C, GPU frequency f_GPU, number of clients K, and number of global rounds G. We solve both problems analytically and empirically using our simulation framework featuring a maximum of 10 Jetson clients with IID Cifar-10 dataset. Our empirical results show that time efficiency is optimal at the maximum available GPU frequency f^(max)_GPU, while energy efficiency is optimal at the highest GPU frequency with minimum supply voltage. Ablation studies reveal that optimal K for time efficiency increases with target accuracy; while optimal K for energy efficiency is always the smallest feasible K. The ablation also shows that optimal C for both time and energy efficiency increase with target accuracy. Based on the optimal hyperparameter setting, we find that optimal communication latency dominates only in low-accuracy regions (20%−29%); whereas optimal compute latency dominates in high-accuracy regions (30%−69%). On the other hand, optimal compute energy dominates across all accuracy regions. These results imply that compute costs are the primary concern for system efficiency under homogeneous FL settings.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2026-02-19 without embargo terms","The student, Jing-Teng Hwang, accepted the attached license on 2025-11-26 at 22:29.","The student, Jing-Teng Hwang, submitted this Thesis for approval on 2025-11-26 at 23:00.","This Thesis was approved for publication on 2025-12-02 at 10:13.","DSpace SAF Submission Ingestion Package generated from Vireo submission #22975 on 2026-02-19 at 18:25:32"],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/132539"],"dc:language":["en"],"dc:rights":["Copyright 2025 Jing-Teng Hwang"],"dc:subject":["Edge Device","Federated Learning"],"dc:title":["Energy and time efficient federated learning"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Electrical & Computer Engr"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:07Z"}