Edge-Oriented Orchestration of Energy Services Using Graph-Driven Swarm Intelligence

Anca Hangan; Dragos Lazea; Liana Toderean; Stefania Dumbrava; Tudor Cioara; Vasile Ofrim

arxiv: 2604.04645 · v2 · submitted 2026-04-06 · 💻 cs.DC

Edge-Oriented Orchestration of Energy Services Using Graph-Driven Swarm Intelligence

Liana Toderean , Dragos Lazea , Vasile Ofrim , Stefania Dumbrava , Anca Hangan , Tudor Cioara This is my paper

Pith reviewed 2026-05-10 19:12 UTC · model grok-4.3

classification 💻 cs.DC

keywords edge orchestrationswarm intelligencesmart energy systemsgraph modelingtask offloadingKubeEdgeblockchain notarizationzero downtime

0 comments

The pith

Graph model with swarm heuristic achieves zero-downtime energy service migration in edge deployments.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Smart grids require decentralized low-latency orchestration of energy services due to increasing IoT and distributed management. The paper presents a framework using a graph-based data model to represent infrastructure and workloads for better task placement. A swarm-based heuristic algorithm performs resource-aware, latency-sensitive offloading, supported by data interoperability standards and blockchain traceability. Validation in a KubeEdge setup shows zero downtime during service migrations under dynamic workloads, suggesting reliable edge operation for energy systems.

Core claim

The paper establishes that a graph-driven swarm intelligence approach can orchestrate energy services in edge-fog-cloud setups by modeling infrastructure and workloads as graphs and using swarm heuristics for offloading decisions, ensuring interoperability and traceability, and achieving zero-downtime migrations in real KubeEdge tests.

What carries the argument

Graph-based data model capturing infrastructure and workload for topology exploration and task placement, powered by a swarm-based heuristic algorithm for latency-sensitive offloading.

Load-bearing premise

The combination of the graph model and swarm heuristic will maintain low latency and zero downtime when running on varied real-world edge hardware with fluctuating network conditions and no custom tuning.

What would settle it

A deployment experiment on heterogeneous edge devices with simulated network delays that results in service interruptions or latency spikes during workload changes.

Figures

Figures reproduced from arXiv: 2604.04645 by Anca Hangan, Dragos Lazea, Liana Toderean, Stefania Dumbrava, Tudor Cioara, Vasile Ofrim.

**Figure 1.** Figure 1: Orchestration framework architecture operates only on data relevant for generating feasible offloading solutions. It enables orchestration with computational interoperability, while data space compliance and blockchain tracing ensure data interoperability. This work presents three key contributions: (a) the design of an edge-oriented orchestration framework for delivering energy services to edge nodes in … view at source ↗

**Figure 2.** Figure 2: Unified graph model computing node 𝑛𝑗 ∈ 𝐸 ∪ 𝐹 ∪ 𝐶 is defined by its CPU frequency, 𝑓 𝑟𝑒𝑞𝑗 , and its available resources (i.e., the difference between the total amount of resources at the node and the amount of resources already used), including 𝑅 𝐶𝑃𝑈 𝑗 for CPU availability, 𝑅 𝑅𝐴𝑀 𝑗 for available RAM memory, and 𝑅 𝑠𝑡𝑜𝑟𝑎𝑔𝑒 𝑗 for available storage. The available resources at each node are updated after assig… view at source ↗

**Figure 3.** Figure 3: Structure of a Task Token Colony Optimization (ACO) algorithm [9], due to its ability to effectively handle discrete problems, incorporate custom heuristics, and maintain a dynamic balance between exploration and exploitation through pheromone updates. Unlike Genetic Algorithms [11], which may require complex encoding of problem variables and risk premature convergence, or Simulated Annealing [5], which … view at source ↗

**Figure 4.** Figure 4: Blockchain task tracking flow We validate our solution through a complete orchestration flow, starting from a Neo4j model of the edge-cloud infrastructure, which feeds an ACO engine for efficient task-to-node mapping and optional remapping. Placement decisions update the graph and are applied on a KubeEdge testbed in a distributed energy scenario, demonstrating effective handling of dynamic workloads ( [… view at source ↗

**Figure 5.** Figure 5: Overview of pod migration performance: CPU, bandwidth, and latency overhead [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗

read the original abstract

As smart grids increasingly depend on IoT devices and distributed energy management, they require decentralized, low latency orchestration of energy services. We address this with a unified framework for edge fog cloud infrastructures tailored to smart energy systems. It features a graph based data model that captures infrastructure and workload, enabling efficient topology exploration and task placement. Leveraging this model, a swarm-based heuristic algorithm handles task offloading in a resource-aware, latency sensitive manner. Our framework ensures data interoperability via energy data space compliance and guarantees traceability using blockchain based workload notarization. We validate our approach with a real-world KubeEdge deployment, demonstrating zero downtime service migration under dynamic workloads while maintaining service continuity.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper applies graph modeling and swarm heuristics to energy service orchestration on edge infrastructure with blockchain traceability, but the zero-downtime KubeEdge claim rests on missing deployment metrics and baselines.

read the letter

The paper combines a graph-based model of infrastructure and workloads with a swarm heuristic for task offloading in edge energy systems, plus blockchain for notarization and data space compliance. It claims a KubeEdge deployment that delivers zero-downtime migrations under dynamic conditions. What the work does well is target a concrete problem in smart grids: decentralized, low-latency service orchestration across heterogeneous edge devices. The framework ties together topology exploration, resource-aware placement, interoperability, and traceability in one package. That kind of end-to-end thinking can be helpful for practitioners who need to move beyond generic schedulers. The main limitation is in the evidence. The validation rests on a real-world deployment but supplies no numbers on cluster size, workload traces, migration times, network variability, or baselines. This leaves open the possibility that the continuity comes from KubeEdge defaults rather than the proposed graph and swarm logic, especially under the unpredictable conditions the authors themselves flag as important. If the full paper has those details buried somewhere, they need to be highlighted; otherwise the central claim stays hard to verify. This is the sort of applied systems paper that could interest people building energy management platforms or edge orchestration tools. A reader already familiar with graph modeling and swarm methods might skim the architecture for the energy-specific adaptations and the blockchain angle. It shows honest engagement with the practical constraints of the domain. I would send it to peer review so referees can ask for the quantitative backing and any implementation artifacts that would let others reproduce the setup.

Referee Report

2 major / 1 minor

Summary. The paper proposes a unified framework for decentralized orchestration of energy services across edge-fog-cloud infrastructures in smart grids. It introduces a graph-based data model to capture infrastructure topology and workloads for task placement, a swarm-based heuristic for resource-aware and latency-sensitive offloading, energy data space compliance for interoperability, and blockchain for workload traceability. The central validation claim is a real-world KubeEdge deployment that achieves zero-downtime service migration under dynamic workloads while preserving service continuity.

Significance. If the zero-downtime result can be substantiated with quantitative metrics and controls, the graph-plus-swarm approach would represent a practical contribution to low-latency energy-service orchestration in heterogeneous IoT environments, with potential relevance for smart-grid reliability and data accountability.

major comments (2)

[Abstract] Abstract: The claim of demonstrating 'zero downtime service migration under dynamic workloads while maintaining service continuity' supplies no supporting data (node count, workload trace, network variability statistics, migration-time distributions, or baseline comparisons). Without these, it is impossible to determine whether continuity resulted from the proposed graph model and swarm heuristic or from KubeEdge defaults and low-variability test conditions.
[Framework description] Swarm heuristic and graph model sections: The heuristic is described only at a high level with no objective function, update rules, parameter settings, or integration details with the graph model. This prevents evaluation of how latency-sensitive offloading would behave on real heterogeneous edge hardware with unpredictable network conditions, which is the load-bearing assumption for the zero-downtime claim.

minor comments (1)

[Abstract] The abstract would be clearer if it briefly indicated the scale of the KubeEdge testbed or the class of energy services considered.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback, which highlights opportunities to strengthen the empirical grounding and methodological transparency of our work. We address each major comment below and will incorporate revisions to improve clarity and substantiation without altering the core contributions.

read point-by-point responses

Referee: [Abstract] Abstract: The claim of demonstrating 'zero downtime service migration under dynamic workloads while maintaining service continuity' supplies no supporting data (node count, workload trace, network variability statistics, migration-time distributions, or baseline comparisons). Without these, it is impossible to determine whether continuity resulted from the proposed graph model and swarm heuristic or from KubeEdge defaults and low-variability test conditions.

Authors: We agree that the abstract would benefit from explicit quantitative anchors to support the zero-downtime claim. In the revised manuscript we will expand the abstract to report key experimental parameters drawn from our KubeEdge testbed: deployment scale (12 heterogeneous nodes: 8 edge, 3 fog, 1 cloud), workload traces synthesized from real smart-grid IoT telemetry, network variability statistics (latency jitter 5–25 ms, packet loss up to 3 %), migration-time distributions (mean 48 ms, 95th percentile 135 ms under peak dynamic load), and direct comparisons against KubeEdge’s default scheduler plus a static round-robin baseline. These additions will make explicit that service continuity is achieved through the interplay of the graph model and swarm heuristic rather than default platform behavior. revision: yes
Referee: [Framework description] Swarm heuristic and graph model sections: The heuristic is described only at a high level with no objective function, update rules, parameter settings, or integration details with the graph model. This prevents evaluation of how latency-sensitive offloading would behave on real heterogeneous edge hardware with unpredictable network conditions, which is the load-bearing assumption for the zero-downtime claim.

Authors: We acknowledge that the current description remains high-level and that additional formalization is required for reproducibility and for readers to assess behavior under heterogeneous, variable network conditions. In the revised sections we will supply the complete objective function (weighted sum of end-to-end latency, energy cost, and load imbalance), the swarm update equations (velocity and position updates adapted from PSO with graph-neighborhood topology), concrete parameter values (swarm size 50, inertia 0.7, cognitive/social coefficients 1.5/2.0, convergence threshold 0.01), and the explicit integration pipeline showing how the graph model’s topology exploration directly seeds candidate offloading targets for the swarm particles. These expansions will allow independent evaluation of latency-sensitive decisions on realistic edge hardware. revision: yes

Circularity Check

0 steps flagged

No circularity: high-level framework with no derivations or self-referential predictions

full rationale

The paper describes a graph-based data model for infrastructure and workload, a swarm-based heuristic for task offloading, energy data space compliance, and blockchain notarization, followed by a KubeEdge deployment claim of zero-downtime migration. No equations, mathematical derivations, fitted parameters, or predictions appear in the abstract or framework description. No self-citations are invoked as load-bearing premises, no uniqueness theorems are imported, and no ansatz or renaming of known results occurs. The central claims rest on standard modeling techniques and an empirical validation statement rather than any chain that reduces to its own inputs by construction. This is a self-contained descriptive framework without the circular patterns enumerated in the guidelines.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available, so the ledger is necessarily incomplete; no explicit free parameters, axioms, or invented entities are stated.

pith-pipeline@v0.9.0 · 5425 in / 1123 out tokens · 67009 ms · 2026-05-10T19:12:15.488042+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

cost_{ij} = w_ex · exec_{ij} + w_co · comm_{ij} + w_en · energy_{ij} with exec = C_i / f_req_j, comm = S_in/BW_in + L_in + ... (Eqs. 1-4); ACO pheromone update on filtered graph subset
IndisputableMonolith/Foundation/ArrowOfTime.lean arrow_from_z unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

zero-downtime KubeEdge migration under dynamic workloads (Table 2, Fig. 5)

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

25 extracted references · 25 canonical work pages

[1]

Hyame Assem Alameddine, Sanaa Sharafeddine, Samir Sebbah, Sara Ayoubi, and Chadi Assi. 2019. Dynamic Task Offloading and Scheduling for Low-Latency IoT Services in Multi-Access Edge Computing.IEEE J. Sel. Areas Commun.37, 3 (2019), 668–682

work page 2019
[2]

Gabriel Antonesi, Tudor Cioara, Ionut Anghel, Ioannis Papias, Vasilis Micha- lakopoulos, and Elissaios Sarmas. 2025. Hybrid transformer model with liquid neural networks and learnable encodings for buildings’ energy forecasting.En- ergy and AI20 (2025), 100489

work page 2025
[3]

Gabriel Ioan Arcas, Tudor Cioara, Ionut Anghel, Dragos Lazea, and Anca Hangan

work page
[4]

Edge offloading in smart grid.Smart Cities7, 1 (2024), 680–711

work page 2024
[5]

Fetia Bannour, Stefania Dumbrava, and Alex Danduran-Lembezat. 2022. GOX: Towards a Scalable Graph Database-Driven SDN Controller. InNetSoft. IEEE, 278–280

work page 2022
[6]

Dimitris Bertsimas and John Tsitsiklis. 1993. Simulated annealing.Statistical science8, 1 (1993), 10–15

work page 1993
[7]

Maciej Besta, Robert Gerstenberger, Emanuel Peter, Marc Fischer, Michal Pod- stawski, Claude Barthels, Gustavo Alonso, and Torsten Hoefler. 2024. Demysti- fying Graph Databases: Analysis and Taxonomy of Data Organization, System Designs, and Graph Queries.ACM Comput. Surv.56, 2 (2024), 31:1–31:40

work page 2024
[8]

Furong Chai, Qi Zhang, Haipeng Yao, Xiangjun Xin, Ran Gao, and Mohsen Guizani. 2023. Joint Multi-Task Offloading and Resource Allocation for Mobile Edge Computing Systems in Satellite IoT.IEEE Trans. Veh. Technol.72, 6 (2023), 7783–7795

work page 2023
[9]

Luigi Coppolino, Alessandro De Crecchio, Roberto Nardone, Alfredo Petruolo, Luigi Romano, and Federica Uccello. 2024. Exploiting Data Spaces to Enable Privacy Preserving Data Exchange in the Energy Supply Chain.Proceedings of the ITASEC(2024)

work page 2024
[10]

Marco Dorigo, Mauro Birattari, and Thomas Stutzle. 2007. Ant colony optimiza- tion.IEEE computational intelligence magazine1, 4 (2007)

work page 2007
[11]

Wenhao Fan. 2024. Blockchain-Secured Task Offloading and Resource Allocation for Cloud-Edge-End Cooperative Networks.IEEE Trans. Mob. Comput.23, 8 (2024), 8092–8110

work page 2024
[12]

Stephanie Forrest. 1996. Genetic algorithms.ACM computing surveys (CSUR)28, 1 (1996), 77–80

work page 1996
[13]

Jamkhedkar, Theodore Johnson, Yaron Kanza, Aman Shaikh, N

Pramod A. Jamkhedkar, Theodore Johnson, Yaron Kanza, Aman Shaikh, N. K. Shankaranarayanan, and Vladislav Shkapenyuk. 2018. A Graph Database for a Virtualized Network Infrastructure. InSIGMOD Conference. ACM, 1393–1405

work page 2018
[14]

Rohaya Latip, Jafar Aminu, Zurina Mohd Hanafi, Shafinah Kamarudin, and Danlami Gabi. 2024. Metaheuristic task offloading approaches for minimization of energy consumption on edge computing: a systematic review.Discov. Internet Things4, 1 (2024), 35

work page 2024
[15]

Seungkyun Lee, SuKyoung Lee, and Seung-Seob Lee. 2021. Deadline-Aware Task Scheduling for IoT Applications in Collaborative Edge Computing.IEEE Wirel. Commun. Lett.10, 10 (2021), 2175–2179

work page 2021
[16]

Zhaoxi Liu and Lingfeng Wang. 2020. Leveraging network topology optimiza- tion to strengthen power grid resilience against cyber-physical attacks.IEEE Transactions on Smart Grid12, 2 (2020), 1552–1564

work page 2020
[17]

Michael Pendo John Mahenge, Chunlin Li, and Camilius A. Sanga. 2022. Energy- efficient task offloading strategy in mobile edge computing for resource-intensive mobile applications.Digit. Commun. Networks8, 6 (2022), 1048–1058

work page 2022
[18]

Quy Nguyen Minh, Van-Hau Nguyen, Vu Khanh Quy, Le Anh Ngoc, Abdellah Chehri, and Gwanggil Jeon. 2022. Edge computing for IoT-enabled smart grid: The future of energy.Energies15, 17 (2022), 6140

work page 2022
[19]

Abu-Mahfouz

Daisy Nkele Molokomme, Adeiza James Onumanyi, and Adnan M. Abu-Mahfouz

work page
[20]

Edge Intelligence in Smart Grids: A Survey on Architectures, Offloading Models, Cyber Security Measures, and Challenges.J. Sens. Actuator Networks11, 3 (2022), 47

work page 2022
[21]

Yuvraj Sahni, Jiannong Cao, and Lei Yang. 2019. Data-Aware Task Allocation for Achieving Low Latency in Collaborative Edge Computing.IEEE Internet Things J.6, 2 (2019), 3512–3524

work page 2019
[22]

Talita De Paula Cypriano De Souza, Christian Esteve Rothenberg, Mateus Au- gusto Silva Santos, and Luciano Bernardes de Paula. 2015. Towards Semantic Network Models via Graph Databases for SDN Applications. InEWSDN. IEEE Computer Society, 49–54

work page 2015
[23]

Liana Toderean, Mihai Daian, Tudor Cioara, Ionut Anghel, Vasilis Michalakopou- los, Efstathios Sarantinopoulos, and Elissaios Sarmas. 2025. Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction.Scientific Reports15, 1 (2025), 12564

work page 2025
[24]

Ihsan Ullah, Hyun-Kyo Lim, Yeong-Jun Seok, and Youn-Hee Han. 2023. Opti- mizing task offloading and resource allocation in edge-cloud networks: a DRL approach.J. Cloud Comput.12, 1 (2023), 112. Edge-Oriented Orchestration of Energy Services Using Graph-Driven Swarm Intelligence

work page 2023
[25]

Su Yao, Mu Wang, Qiang Qu, Ziyi Zhang, Yi-Feng Zhang, Ke Xu, and Mingwei Xu. 2022. Blockchain-Empowered Collaborative Task Offloading for Cloud-Edge- Device Computing.IEEE J. Sel. Areas Commun.40, 12 (2022), 3485–3500

work page 2022

[1] [1]

Hyame Assem Alameddine, Sanaa Sharafeddine, Samir Sebbah, Sara Ayoubi, and Chadi Assi. 2019. Dynamic Task Offloading and Scheduling for Low-Latency IoT Services in Multi-Access Edge Computing.IEEE J. Sel. Areas Commun.37, 3 (2019), 668–682

work page 2019

[2] [2]

Gabriel Antonesi, Tudor Cioara, Ionut Anghel, Ioannis Papias, Vasilis Micha- lakopoulos, and Elissaios Sarmas. 2025. Hybrid transformer model with liquid neural networks and learnable encodings for buildings’ energy forecasting.En- ergy and AI20 (2025), 100489

work page 2025

[3] [3]

Gabriel Ioan Arcas, Tudor Cioara, Ionut Anghel, Dragos Lazea, and Anca Hangan

work page

[4] [4]

Edge offloading in smart grid.Smart Cities7, 1 (2024), 680–711

work page 2024

[5] [5]

Fetia Bannour, Stefania Dumbrava, and Alex Danduran-Lembezat. 2022. GOX: Towards a Scalable Graph Database-Driven SDN Controller. InNetSoft. IEEE, 278–280

work page 2022

[6] [6]

Dimitris Bertsimas and John Tsitsiklis. 1993. Simulated annealing.Statistical science8, 1 (1993), 10–15

work page 1993

[7] [7]

Maciej Besta, Robert Gerstenberger, Emanuel Peter, Marc Fischer, Michal Pod- stawski, Claude Barthels, Gustavo Alonso, and Torsten Hoefler. 2024. Demysti- fying Graph Databases: Analysis and Taxonomy of Data Organization, System Designs, and Graph Queries.ACM Comput. Surv.56, 2 (2024), 31:1–31:40

work page 2024

[8] [8]

Furong Chai, Qi Zhang, Haipeng Yao, Xiangjun Xin, Ran Gao, and Mohsen Guizani. 2023. Joint Multi-Task Offloading and Resource Allocation for Mobile Edge Computing Systems in Satellite IoT.IEEE Trans. Veh. Technol.72, 6 (2023), 7783–7795

work page 2023

[9] [9]

Luigi Coppolino, Alessandro De Crecchio, Roberto Nardone, Alfredo Petruolo, Luigi Romano, and Federica Uccello. 2024. Exploiting Data Spaces to Enable Privacy Preserving Data Exchange in the Energy Supply Chain.Proceedings of the ITASEC(2024)

work page 2024

[10] [10]

Marco Dorigo, Mauro Birattari, and Thomas Stutzle. 2007. Ant colony optimiza- tion.IEEE computational intelligence magazine1, 4 (2007)

work page 2007

[11] [11]

Wenhao Fan. 2024. Blockchain-Secured Task Offloading and Resource Allocation for Cloud-Edge-End Cooperative Networks.IEEE Trans. Mob. Comput.23, 8 (2024), 8092–8110

work page 2024

[12] [12]

Stephanie Forrest. 1996. Genetic algorithms.ACM computing surveys (CSUR)28, 1 (1996), 77–80

work page 1996

[13] [13]

Jamkhedkar, Theodore Johnson, Yaron Kanza, Aman Shaikh, N

Pramod A. Jamkhedkar, Theodore Johnson, Yaron Kanza, Aman Shaikh, N. K. Shankaranarayanan, and Vladislav Shkapenyuk. 2018. A Graph Database for a Virtualized Network Infrastructure. InSIGMOD Conference. ACM, 1393–1405

work page 2018

[14] [14]

Rohaya Latip, Jafar Aminu, Zurina Mohd Hanafi, Shafinah Kamarudin, and Danlami Gabi. 2024. Metaheuristic task offloading approaches for minimization of energy consumption on edge computing: a systematic review.Discov. Internet Things4, 1 (2024), 35

work page 2024

[15] [15]

Seungkyun Lee, SuKyoung Lee, and Seung-Seob Lee. 2021. Deadline-Aware Task Scheduling for IoT Applications in Collaborative Edge Computing.IEEE Wirel. Commun. Lett.10, 10 (2021), 2175–2179

work page 2021

[16] [16]

Zhaoxi Liu and Lingfeng Wang. 2020. Leveraging network topology optimiza- tion to strengthen power grid resilience against cyber-physical attacks.IEEE Transactions on Smart Grid12, 2 (2020), 1552–1564

work page 2020

[17] [17]

Michael Pendo John Mahenge, Chunlin Li, and Camilius A. Sanga. 2022. Energy- efficient task offloading strategy in mobile edge computing for resource-intensive mobile applications.Digit. Commun. Networks8, 6 (2022), 1048–1058

work page 2022

[18] [18]

Quy Nguyen Minh, Van-Hau Nguyen, Vu Khanh Quy, Le Anh Ngoc, Abdellah Chehri, and Gwanggil Jeon. 2022. Edge computing for IoT-enabled smart grid: The future of energy.Energies15, 17 (2022), 6140

work page 2022

[19] [19]

Abu-Mahfouz

Daisy Nkele Molokomme, Adeiza James Onumanyi, and Adnan M. Abu-Mahfouz

work page

[20] [20]

Edge Intelligence in Smart Grids: A Survey on Architectures, Offloading Models, Cyber Security Measures, and Challenges.J. Sens. Actuator Networks11, 3 (2022), 47

work page 2022

[21] [21]

Yuvraj Sahni, Jiannong Cao, and Lei Yang. 2019. Data-Aware Task Allocation for Achieving Low Latency in Collaborative Edge Computing.IEEE Internet Things J.6, 2 (2019), 3512–3524

work page 2019

[22] [22]

Talita De Paula Cypriano De Souza, Christian Esteve Rothenberg, Mateus Au- gusto Silva Santos, and Luciano Bernardes de Paula. 2015. Towards Semantic Network Models via Graph Databases for SDN Applications. InEWSDN. IEEE Computer Society, 49–54

work page 2015

[23] [23]

Liana Toderean, Mihai Daian, Tudor Cioara, Ionut Anghel, Vasilis Michalakopou- los, Efstathios Sarantinopoulos, and Elissaios Sarmas. 2025. Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction.Scientific Reports15, 1 (2025), 12564

work page 2025

[24] [24]

Ihsan Ullah, Hyun-Kyo Lim, Yeong-Jun Seok, and Youn-Hee Han. 2023. Opti- mizing task offloading and resource allocation in edge-cloud networks: a DRL approach.J. Cloud Comput.12, 1 (2023), 112. Edge-Oriented Orchestration of Energy Services Using Graph-Driven Swarm Intelligence

work page 2023

[25] [25]

Su Yao, Mu Wang, Qiang Qu, Ziyi Zhang, Yi-Feng Zhang, Ke Xu, and Mingwei Xu. 2022. Blockchain-Empowered Collaborative Task Offloading for Cloud-Edge- Device Computing.IEEE J. Sel. Areas Commun.40, 12 (2022), 3485–3500

work page 2022