International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 04 | Apr 2026 www.irjet.net p-ISSN: 2395-0072
COMPARATIVE STUDY OF EDGE COMPUTING NETWORKS Vs. CLOUD COMPUTING NETWORKS FOR LATENCY-SENSITIVE APPLICATIONS
Kajal Singh1 , Mrs. Sahreen Hijab2
1Master of Technology, Computer Science and Engineering, Sagar Institute of Technology and Management, Barabanki, India
2Assistant Professor, Department of Computer Science and Engineering, Sagar Institute of Technology and Management, Barabanki, India
Abstract - The rapid growth of latency-sensitive applications such as Internet of Things (IoT) systems, autonomous vehicles, augmented reality, and industrial automationhasexposedsignificantlimitationsintraditional cloudcomputing networks. Although cloudcomputingoffers high scalability and resource elasticity, its centralized architecture introduces considerable communication delays, making it less suitable for real-time applications requiring strict Quality of Service (QoS) guarantees. To address these challenges, edge computing has emerged as a decentralized paradigm that processes data closer to the source, thereby reducing latency and improving responsiveness. This paper presents a comprehensive comparative study of edge computing networks and cloud computing networks for latency-sensitiveapplications.Ahybridresearchmethodology is adopted, integrating analytical latency modeling and simulation-based evaluation using EdgeCloudSim. Key performance metrics, including end-to-end latency, jitter, response time, bandwidth utilization, and scalability, are systematicallyanalyzedacrossmultiplereal-worldscenarios such as IoT monitoring, autonomous systems, augmented reality, and industrial automation. The results demonstrate that edge computing significantly outperforms cloud computing in terms of latency and real-time responsiveness, particularlyinscenariosrequiringmillisecond-leveldecisionmaking. However, cloud computing continues to provide advantages in scalability and centralized resource management. The study concludes that a hybrid edge–cloud architecture offers an optimal solution by balancing lowlatencyperformance withscalablecomputingresources.
Key Words: Edge Computing, Cloud Computing, Latency-Sensitive Applications, Quality of Service (QoS), EdgeCloudSim, Distributed Systems
1. INTRODUCTION
Therapidevolutionofdistributedcomputingparadigmshas significantly transformed how modern applications are designedanddeployed.Withtheincreasingdemandforrealtime processing and instantaneous decision-making, traditionalcomputingarchitecturesarebeingre-evaluated to meet stringent latency and Quality of Service (QoS) requirements. This section introduces the background, problemcontext,researchgap,contributions,andstructure
ofthestudy,focusingonthe comparativeanalysisofedge and cloud computing networks for latency-sensitive applications.
1.1 Background
Theprogressionofcomputingparadigmshasevolvedfrom centralizedmainframesystemstodistributedarchitectures, followed by the emergence of cloud computing and, more recently,edgecomputing.Earlycentralizedsystemsoffered limited scalability and flexibility, which led to the development of distributed computing models capable of sharingresourcesacrossmultiplenodes.Cloudcomputing further advanced this paradigm by enabling on-demand access to scalable and virtualized resources through centralized data centers. However, as application requirementsshiftedtowardreal-timeresponsiveness,the limitations of cloud-centric models became evident, particularly in terms of latency and network dependency (Buyyaetal.,2009).
1.1.1 Evolution from Centralized to Edge Computing
Thetransitionfromcentralizedtoedgecomputingreflects the need to minimize communication delays and enhance systemresponsiveness.Whilecloudcomputingcentralizes processing in distant data centers, edge computing distributescomputationalresourcesclosertoendusersand datasources.Thisproximityreducespropagationdelayand networkcongestion,makingedgecomputingmoresuitable for latency-critical environments. The integration of edge nodes such as gateways and micro data centers enables localized processing, thereby improving performance for time-sensitiveapplications(Satyanarayanan,2017).
1.1.1.1 Rise of Real-Time Applications
The proliferation of real-time applications has been a key driverbehindtheadoptionofedgecomputing.Technologies such as Internet of Things (IoT), augmented reality (AR), virtualreality(VR),andautonomoussystemsrequirerapid dataprocessingandnear-instantaneousresponsetimes.For instance, autonomous vehicles rely on millisecond-level decision-making,whileAR/VRapplicationsdemandultralowlatencytomaintainimmersiveuserexperiences.These requirements cannot be consistently met by centralized
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 04 | Apr 2026 www.irjet.net p-ISSN: 2395-0072
cloud infrastructures, thereby necessitating decentralized computingapproaches(Shietal.,2016).
1.2 Problem Statement
Despite its advantages in scalability and resource management,cloudcomputingfacesinherentchallengesin supporting latency-sensitive applications due to its centralizedarchitecture.Datageneratedatenddevicesmust travellongdistancestoreachclouddatacenters,resultingin increasedtransmissionandpropagationdelays.Additionally, networkcongestionandvariabilityinroutingpathsfurther exacerbate latency issues, leading to inconsistent performanceinreal-timescenarios.
Thecoreproblemaddressedinthisresearchistheinability oftraditionalcloudcomputingnetworkstoguaranteelowlatency performance for time-critical applications. While edge computing offers a potential solution by bringing computationclosertodatasources,itseffectivenessmustbe systematically evaluated. Therefore, there is a need for a comprehensive comparative analysis of edge and cloud computingnetworksunderrealisticconditionstodetermine theirsuitabilityforlatency-sensitiveapplications(Machand Becvar,2017).
1.3 Research Gap
Althoughextensiveresearchhasbeenconductedoncloud andedgecomputingindependently,thereremainsalackof standardized comparative studies focusing specifically on latencyperformance.Many existingworksevaluateeither cloud or edge computing in isolation, without providing a unified framework for comparison across diverse applicationscenarios.Furthermore,thereislimiteduse of integrated analytical and simulation-based approaches to validate performance differences under controlled experimentalconditions.
Another significant gap lies in the absence of crossapplicationanalysis.Moststudiesfocusonasingledomain, suchasIoTorautonomoussystems,withoutconsideringthe variability in latency requirements across different applications.Thislimitationrestrictsthegeneralizabilityof findings and highlights the need for a comprehensive evaluationframeworkthatincorporatesmultiplereal-world scenarios and standardized performance metrics (Chiang andZhang,2016).
2. RELATED WORK
Therapidevolutionofdistributedcomputingparadigmshas ledtoextensiveresearchonbothcloudandedgecomputing, particularlyinthecontextofperformanceoptimizationfor latency-sensitiveapplications.Thissectionreviewsexisting studies on cloud computing performance, recent advancements in edge computing, and prior comparative
analyses,followedbyasynthesisoftheresearchgapsthat motivatethepresentstudy.
2.1 Cloud Computing Performance Studies
Cloud computing has been widely studied as a dominant paradigmforscalableandon-demandresourceprovisioning. Earlyresearchprimarilyfocusedonitsabilitytodeliverhigh computational power, storage capacity, and elasticity through centralized data centers. However, performance evaluations have increasingly highlighted limitations in handlinglatency-sensitiveworkloads.
2.1.1 Latency Challenges in Cloud Networks
Latency in cloud computing environments arises from multiplesources,includinglong-distancedatatransmission, networkcongestion,andqueuingdelayswithindatacenters. Asuserrequestsmusttraversewide-areanetworks(WANs) to reach centralized servers, the physical distance significantlycontributestopropagationdelay.Studieshave shownthatthisdelaybecomesacriticalbottleneckforrealtimeapplications,particularlywhenconsistentlow-latency performanceisrequired(Armbrustetal.,2010).
2.1.1.1 Scalability and Bandwidth Considerations
Whilecloudcomputingexcelsinscalabilitythroughelastic resourceallocation,thisadvantageoftencomesatthecostof increased bandwidth consumption and network dependency. High volumes of data generated by modern applications place significant strain on network infrastructure, leading to congestion and reduced throughput.Althoughtechniquessuchasloadbalancingand virtualizationimproveefficiency,theydonotfullyeliminate latency variability caused by bandwidth limitations and dynamictrafficpatterns(Zhangetal.,2010).
2.2 Edge Computing Advancements
Edgecomputinghasemergedasacomplementaryparadigm designed to address the limitations of cloud-centric architectures. By decentralizing computation and placing resources closer to end users, edge computing reduces communication delays and enhances real-time responsiveness.
2.2.1 Evolution of Edge, Fog, and Cloudlet Paradigms
Thedevelopmentofedgecomputinghasbeeninfluencedby relatedparadigmssuchasfogcomputingandcloudlets.Fog computing extends cloud capabilities to intermediate network layers, enabling distributed processing across routers and gateways. Cloudlets, on the other hand, representsmall-scaledatacenterslocatednearmobileusers, providing localized computational resources. These paradigms collectively contribute to the evolution of edge computingbyemphasizingproximity-basedprocessingand reducedlatency(Bonomietal.,2012).
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 04 | Apr 2026 www.irjet.net p-ISSN: 2395-0072
2.2.1.1 Role of Edge Computing in Low-Latency Systems
Edgecomputingplaysacriticalroleinsupportinglatencysensitiveapplicationsbyenablingreal-timedataprocessing at the network periphery. By minimizing the distance betweendatasourcesandprocessingnodes,edgecomputing significantly reduces end-to-end latency and improves system responsiveness. This approach is particularly effectiveinapplicationssuchasIoT,autonomoussystems, and augmented reality, where rapid decision-making is essential. Furthermore, edge computing reduces backhaul traffic and enhances system reliability by allowing partial operationduringnetworkdisruptions(Shietal.,2016).
2.3 Comparative Studies
Severalstudieshaveattemptedtocomparecloudandedge computing paradigms to evaluate their respective performance characteristics. These comparisons typically focusonmetricssuchaslatency,bandwidthutilization,and computationalefficiency
2.3.1 Existing Comparative Analyses
Existingcomparativestudiesgenerallyconcludethatedge computingoutperformscloudcomputinginlatency-sensitive scenarios due to its proximity to end devices. Conversely, cloudcomputingisrecognizedforitssuperiorscalabilityand centralized resource management capabilities. Some research has also explored hybrid architectures that combineedgeandcloudresourcestobalanceperformance andscalability(Satyanarayanan,2017).
2.3.1.1
Limitations of Existing Studies
Despitetheseinsights,existingcomparativestudiesexhibit severallimitations.Manyanalysesrelyonsimplifiedmodels or theoretical assumptions without validating results through simulation or real-world experimentation. Additionally, most studies focus on a single application domain,limitingthegeneralapplicabilityoftheirfindings. The lack of standardized evaluation frameworks and consistent performance metrics further complicates the comparisonofresultsacrossdifferentstudies,highlighting theneedformorecomprehensiveandsystematicresearch (MachandBecvar,2017).
3. SYSTEM MODEL AND ARCHITECTURE
The system model defines the structural and operational framework used to analyze and compare cloud and edge computing networks. It captures how computational resources are organized, how data flows through the network, and how tasks are processed under different architectural paradigms. This section presents the cloud computingmodel,edgecomputingmodel,andhybridedge–cloud architecture, followed by a comparative framework highlightingtheirkeydifferences.
3.1 Cloud Computing Network Model
The cloud computing network model is based on a centralizedarchitectureinwhichcomputationalresources, storage, and services are hosted within large-scale data centers. These data centers are typically located at geographically distant locations and are interconnected throughhigh-speedbackbonenetworks.End-userdevices, such as mobile phones, sensors, and computers, send requests to these centralized servers for processing and receiveresponsesoverthenetwork.
3.1.1 Centralized Data Centers
Centralizeddatacentersformthecoreofcloudcomputing infrastructure, providing high computational capacity, storage scalability, and virtualization capabilities. These facilitieshostthousandsofserversthatoperateinparallelto handle large volumes of user requests. The centralized nature of cloud computing enables efficient resource management,loadbalancing,andelasticscaling.However, this centralization also introduces dependency on remote infrastructure, which can impact performance for timesensitiveapplications.
3.1.1.1
WAN-Based Communication
Communicationincloudcomputingenvironmentsprimarily occurs over wide-area networks (WANs). User requests must travel through multiple network layers, including accessnetworks,corenetworks,anddatacenternetworks, before reaching the cloud servers. This multi-hop communication increases propagation and transmission delays, leading to higher end-to-end latency. Additionally, WAN-basedcommunicationissusceptibletocongestionand bandwidth limitations, which can further degrade performanceinlatency-criticalscenarios.
3.2 Edge Computing Network Model
Theedgecomputingnetworkmodeladoptsadecentralized approachbydistributingcomputationalresourcescloserto end users and data sources. Instead of relying solely on distant cloud servers, edge computing introduces intermediatenodesthatperformdataprocessingatornear the network edge. This model is particularly effective for applications requiring real-time responsiveness and lowlatencycommunication.
3.2.1 Distributed Edge Nodes
Edge computing relies on a network of distributed nodes, such as gateways, routers, base stations, and micro data centers, equipped with computational and storage capabilities.Thesenodesarestrategicallyplacednearend devicestoreducethephysicaldistancethatdatamusttravel. Byprocessingdatalocally,edgenodesminimizerelianceon centralized infrastructure and improve system efficiency. However,theirlimitedcomputationalcapacitycomparedto
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 04 | Apr 2026 www.irjet.net p-ISSN: 2395-0072
cloud data centers presents challenges in handling largescaleworkloads.
3.2.1.1 Role of Gateways and Micro Data Centers
Gatewaysandmicrodatacentersserveaskeycomponents in edge computing architectures. Gateways act as intermediaries between end devices and higher-level networks, performing functions such as data aggregation, protocoltranslation,andpreliminaryanalytics.Microdata centers provide localized computing power and storage, enablingmorecomplexprocessingtaskstobeexecutednear thedatasource.Together,thesecomponentssupportrealtime decision-making and reduce the need for continuous datatransmissiontothecloud.
3.3 Hybrid Edge–Cloud Architecture
Thehybridedge–cloudarchitecturecombinesthestrengths of both cloud and edge computing to achieve a balance between low latency and high scalability. In this model, computational tasks are dynamically distributed between edge nodes and cloud data centers based on application requirements,networkconditions,andresourceavailability.
3.3.1 Task Partitioning Approach
Task partitioning is a key mechanism in hybrid architectures, where workloads are divided into latencysensitive and compute-intensive components. Latencycritical tasks are executed at the edge to ensure rapid response times, while computationally intensive or nontime-critical tasks are offloaded to the cloud for efficient processing.Thisapproachoptimizesresourceutilizationand reducesnetworkcongestionwhilemaintainingperformance forreal-timeapplications.
3.3.1.1 Dynamic Offloading and Coordination
Dynamic offloading strategies enable the system to adaptivelydecidewheretasksshouldbeprocessed.Factors suchasnetworklatency,nodecapacity,workloadintensity, andapplicationpriorityinfluencethesedecisions.Effective coordinationbetweenedgeandcloudcomponentsensures seamless data flow and consistent service delivery. This hybrid approach is increasingly adopted in modern distributed systems to address the limitations of purely centralizedordecentralizedmodels.
3.4 Architectural Comparison Framework
A comparative analysis of cloud and edge computing architectures highlights their fundamental differences in termsofresourceplacement,performancecharacteristics, and processing models. This framework provides a structuredbasisforevaluatingtheirsuitabilityfordifferent applicationscenarios.
4. RESEARCH METHODOLOGY
Theresearchmethodologydefinesthestructuredapproach usedtoinvestigateandcomparetheperformanceofcloud and edge computing networks for latency-sensitive applications.Itintegratestheoreticalanalysis,mathematical modeling, simulation-based experimentation, and comparative evaluation to ensure comprehensive and reliableresults.
4.1 Research Design
The research design follows a systematic multi-phase approach to ensure logical progression from conceptual understanding to empirical validation. Each phase contributestobuildingacompleteevaluationframeworkfor analyzinglatencybehavior.
4.2 Simulation Environment
Simulationisusedastheprimaryexperimentalmethodto evaluate system performance under controlled and reproducibleconditions.Itallowsthemodelingofcomplex distributed environments without requiring physical infrastructure.
4.2.1 Simulation Tool: EdgeCloudSim
ThestudyemploysEdgeCloudSim,asimulationframework specifically designed for modeling hybrid cloud-edge computing environments. It extends the capabilities of CloudSim by incorporating edge nodes, mobility, and networkdynamics.
4.2.1.1 Justification for Tool Selection
EdgeCloudSimisselectedduetoitsabilitytosimulateboth centralized cloud and decentralized edge architectures within a unified environment. It supports customizable network topologies, edge node placement, and workload distribution, making it highly suitable for comparative studies. Additionally, it enables detailed measurement of latency,bandwidthusage,andprocessingdelays,whichare criticalforevaluatinglatency-sensitiveapplications.
4.3 Network Configuration
Thenetworkconfigurationdefinesthestructuralparameters of the simulation environment, including distances, bandwidth,andprocessing capabilities.These parameters are designed to reflect realistic distributed computing scenarios.
4.3.1 Communication Distance and Bandwidth
Incloudcomputingscenarios,dataistransmittedoverlong distances to centralized data centers, whereas edge computingprocessesdataclosertothesource.Bandwidth
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 04 | Apr 2026 www.irjet.net p-ISSN: 2395-0072
availability also varies across network layers, influencing transmissiondelay.
Table-1: Network Parameters
Parameter
Distance 500–2000 km 10–50km Affects propagation delay
Bandwidth 10Mbps–10Gbps 10Mbps–1 Gbps Influences transmission delay
Network Type WideArea Network Local/Access Network Determines latency variability
Processing Location Centralized Distributed Impacts responsetime
4.4 Workload Modeling
Workloadmodelingrepresentsthetypesofapplicationsand traffic patterns used in the simulation. It is essential for evaluating how different computing paradigms perform underreal-worldconditions.
Table-2: Workload Characteristics
Application
IoTSystems Periodic 10–100ms Smalldata packets
Autonomou
5. RESULTS AND ANALYSIS
Thissectionpresentstheoutcomesofthesimulation-based evaluation and analytical modeling, focusing on the comparative performance of cloud and edge computing networks across multiple latency-sensitive scenarios. The analysis is structured around key performance metrics,
including latency, jitter, bandwidth utilization, and scalability,followedbyapplication-specificinsights.
5.1 Latency Comparison
Latency is the most critical metric for evaluating the performanceofdistributedcomputingsystems,particularly forreal-timeapplications.Thecomparativeanalysisreveals significantdifferencesbetweencloudandedgecomputing architecturesacrossallexperimentalscenarios.
5.1.1 Edge vs Cloud Across Scenarios
Simulationresultsindicatethatedgecomputingconsistently achieves lower end-to-end latency compared to cloud computing.Thisimprovementisprimarilyduetoreduced propagation and transmission delays, as edge nodes are located closer to end users. In contrast, cloud computing suffers from higher latency due to long-distance communicationandnetworkcongestion.
Table-3: Latency Comparison Table
Scenario Type Cloud
5.2 Jitter and Stability Analysis
Jitter, defined as the variation in latency over time, is a crucialindicatorofsystemstability.Stablesystemsexhibit minimal fluctuations in delay, ensuring consistent performance.
5.2.1 Variation in Latency
The analysis shows that cloud computing environments experiencehigherjitterduetovariablenetworkconditions, including congestion and routing changes in wide-area networks. Edge computing, on the other hand, maintains lower and more consistent latency due to localized processingandreduceddependencyonlongcommunication paths.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 04 | Apr 2026 www.irjet.net p-ISSN: 2395-0072
Table-4: Jitter Comparison Table
CloudComputing 20–60
EdgeComputing 5–15
HybridModel 10–25
6.CONCLUSION
Thisresearchpresentsacomprehensivecomparativestudy of edge computing and cloud computing networks with a focusonlatency-sensitiveapplications.Thestudyintegrates analytical modeling and simulation-based evaluation to examine key performance metrics, including end-to-end latency, jitter, bandwidth utilization, scalability, and responsetime.Thefindingsclearlydemonstratethatedge computing significantly outperforms cloud computing in terms of latency and stability due to its decentralized architecture and proximity to end users. This makes edge computinghighlysuitableforreal-timeapplicationssuchas IoTsystems,autonomousvehicles,augmentedreality,and industrial automation, where rapid decision-making is critical.
However, cloud computing continues to offer substantial advantages in terms of scalability, centralized resource management, and the ability to handle large-scale data processingtasks.Theanalysisalsohighlightsafundamental trade-off between latency and scalability, indicating that neither paradigm alone can fully satisfy the diverse requirementsofmodernapplications.
To address this limitation, the study emphasizes the importance of hybrid edge–cloud architectures, which combinethelow-latencybenefitsofedgecomputingwiththe scalabilityofcloudcomputing.Byleveragingintelligenttask partitioning and dynamic workload distribution, such architectures can achieve optimal system performance. Overall, this research provides valuable insights and a structured framework for designing efficient computing infrastructurestailoredtolatency-sensitiveenvironments.
7.FUTURE SCOPE OF RESEARCH
Future research can extend this work by incorporating emergingtechnologiessuchas5Gand6Gnetworks,which canfurtherenhanceedgecomputingcapabilitiesandreduce communication latency. Additionally, integrating artificial intelligence and machinelearning techniques for dynamic taskoffloadingandresourceallocationcanimprovesystem efficiency. Real-world deployment and experimental validation using physical testbeds would provide more practicalinsightsbeyondsimulation-basedanalysis.Further studies may also explore energy efficiency, security
challenges, and fault tolerance in hybrid edge–cloud environments. Expanding the analysis to include diverse applicationdomainsandheterogeneousnetworkconditions canenhancethegeneralizabilityofthefindings.
REFERENCES
1. Armbrust,M.,Fox,A.,Griffith,R.,Joseph,A.D.,Katz,R., Konwinski,A.,Lee,G.,Patterson,D.,Rabkin,A.,Stoica,I. and Zaharia, M., 2010. A view of cloud computing. CommunicationsoftheACM,53(4),pp.50–58.
2. Bonomi,F.,Milito,R.,Zhu,J.andAddepalli,S.,2012.Fog computing and its role in the internet of things. In: ProceedingsoftheFirstEditionoftheMCCWorkshop onMobileCloudComputing.ACM,pp.13–16.
3. Buyya, R., Yeo, C.S., Venugopal, S., Broberg, J. and Brandic, I., 2009. Cloud computing and emerging IT platforms: Vision, hype, and reality for delivering computing as the 5th utility. Future Generation ComputerSystems,25(6),pp.599–616.
4. Chiang, M. and Zhang, T., 2016. Fog and IoT: An overview of research opportunities. IEEE Internet of ThingsJournal,3(6),pp.854–864.
5. Mach,P.andBecvar,Z.,2017.Mobileedgecomputing:A survey on architecture and computation offloading. IEEE Communications Surveys & Tutorials, 19(3), pp.1628–1656.
6. Satyanarayanan, M., 2017. The emergence of edge computing.Computer,50(1),pp.30–39.
7. Shi, W.,Cao, J.,Zhang, Q., Li, Y.andXu,L.,2016.Edge computing: Vision and challenges. IEEE Internet of ThingsJournal,3(5),pp.637–646.
8. Zhang,Q.,Chen,M.,Li,L.,Zhao,W.andLi,J.,2010.Cloud computing: state-of-the-art and research challenges. Journal of Internet Services and Applications, 1(1), pp.7–18.
9. Andriulo,F.C.,Fiore,M.,Mongiello,M.,Traversa,E.and Zizzo,V.,2024.Edgecomputingandcloudcomputing forinternetofthings:Areview.Informatics,11(4),p.71.
10. Singh,V.K.,Mehra,S.andAli,A.,2024.Edgecomputing integration with 5G for IoT: Framework, challenges, andfuturedirections.ProceedingsofICICC2024.
11. Rana,N.,Jeribi,F.,Khan,Z.,Alrawagfeh,W.,BenDhaou, I.,Haseebuddin,M.andUddin,M.,2024.Asystematic literature review on virtual machine scheduling in cloud and multi-access computing. Frontiers in ComputerScience,6,p.1288552.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 04 | Apr 2026 www.irjet.net p-ISSN: 2395-0072
12. Kawalpreet, Salaria, A., Kumar, J. and Kaur, A., 2024. Edgecomputingtechnologies:Acomprehensivereview andfutureperspectives.AIPConferenceProceedings, 3121,p.040006.
13. SudhakarBabu,S.,2024.Exploringtheroleofedgeand cloud computing in enhancing data analytics for IoT ecosystems. Journal of Recent Trends in Computer ScienceandEngineering,12(1),pp.27–33.
14. Hossain, M.I., Sumon, S.A., Hasan, H.M., Akter, F., Badhon,M.B. andIslam,M.N.U.,2024. Quantum-edge cloud computing: A future paradigm for IoT applications.arXivpreprintarXiv:2405.04824.
15. Azad,M.W.A.,Shannigrahi,S.,Stergiou,N.,Ortega,F.R. andMastorakis,S.,2021.CLEDGE:Ahybridcloud-edge computing framework. arXiv preprint arXiv:2107.07604.
16. Rosendo, D., Costan, A., Valduriez, P. and Antoniu, G., 2022. Distributed intelligence on the edge-to-cloud continuum: A systematic review. Journal of Cloud Computing,11(1),pp.1–25.
17. Varshney,P.andSimmhan,Y.,2017.Demystifyingfog computing:Architecturesandapplications.IEEECloud Computing,4(1),pp.36–44.
18. Chen, J., Ran, X. and Li, K., 2018. Deep learning with edge computing: A review. Proceedings of the IEEE, 106(5),pp.874–893.
19. Satyanarayanan, M., 2017. The emergence of edge computing.Computer,50(1),pp.30–39.
20. Shi, W.,Cao, J.,Zhang, Q., Li, Y.andXu,L.,2016.Edge computing: Vision and challenges. IEEE Internet of ThingsJournal,3(5),pp.637–646.
21. Mach,P.andBecvar,Z.,2017.Mobileedgecomputing:A survey on computation offloading. IEEE CommunicationsSurveys&Tutorials,19(3),pp.1628–1656.
22. Zhang,Q.,Chen,M.,Li,L.,Zhao,W.andLi,J.,2010.Cloud computing: State-of-the-art and research challenges. Journal of Internet Services and Applications, 1(1), pp.7–18.
23. Gubbi, J., Buyya, R., Marusic, S. and Palaniswami, M., 2013. Internet of Things (IoT): A vision and future directions. Future Generation Computer Systems, 29(7),pp.1645–1660.
24. Santos,R.,Gomes,P.anddaSilva,A.,2021.Improving IoTsystemswithedgecomputing:Asystematicreview. JournalofSystemsandSoftware,175,p.110903.
25. Yi,S.,Li,C.andLi,Q.,2015.Asurveyoffogcomputing: Concepts and applications. Proceedings of the WorkshoponMobileBigData.
26. Hong,K.,Lillethun,D.,Ramachandran,U.,Ottenwälder, B.andKoldehofe,B.,2013.Mobilefog:Aprogramming model for large-scale applications. ACM SIGCOMM Workshop.
27. Dinh, H.T., Lee, C., Niyato, D. and Wang, P., 2013. A survey of mobile cloud computing. IEEE CommunicationsSurveys&Tutorials,15(3),pp.1347–1370.
28. Taleb,T.,Samdanis,K.,Mada,B.,Flinck,H.,Dutta,S.and Sabella, D., 2017. On multi-access edge computing: A survey. IEEE Communications Surveys & Tutorials, 19(3),pp.1657–1681.
29. Zhou,Z.,Chen,X.,Li,E.,Zeng,L.,Luo,K.andZhang,J., 2019.Edgeintelligence:PavingthelastmileofAIwith edge computing. Proceedings of the IEEE, 107(8), pp.1738–1762.
30. Li, W., Zhang, Y., Zhou, Y. and Li, Y., 2021. Edge computingforautonomousdriving:Opportunitiesand challenges.IEEENetwork,35(5),pp.146–153.
31. Abbas,N.,Zhang,Y.,Taherkordi,A.andSkeie,T.,2018. Mobile edge computing: A survey. IEEE Internet of ThingsJournal,5(1),pp.450–465.
32. Rahmani, A.M., Gia, T.N., Negash, B., Anzanpour, A., Azimi, I., Jiang, M. and Liljeberg, P., 2018. Exploiting smarte-healthgatewaysattheedgeofhealthcareIoT. FutureGenerationComputerSystems,78,pp.641–658.
33. Zhou,F.,Hu,R.Q.,Li,Z.andWang,Y.,2019.Mobileedge computing in 5G networks: A survey. China Communications,16(3),pp.1–17.