Performance Evaluation of Hybrid Cloud-Fog Computing Architectures in Smart Home IoT Environments: A Comparative Simulation Study Across Multiple Tools

This study evaluates a hybrid cloud-fog computing architecture within a smart home environment consisting of 50 IoT devices. The performance is assessed using seven simulation tools: MATLAB, iFogSim, EdgeCloudSim, YAFS, FogNetSim++, PureEdgeSim, and LEAF. Simulation results demonstrate that distribu...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of grid computing Ročník 23; číslo 2; s. 14
Hlavní autoři: Ruchika, Chhillar, Rajender Singh
Médium: Journal Article
Jazyk:angličtina
Vydáno: Dordrecht Springer Nature B.V 01.06.2025
Témata:
Central processing units
CPUs
Edge computing
Energy consumption
Energy efficiency
Home environment
Internet of Things
IoT devices
Matlab
Memory tasks
Nodes
Performance evaluation
Resource allocation
Resource efficiency
Simulation
Smart buildings
Smart houses
Utilization
ISSN:1570-7873, 1572-9184
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This study evaluates a hybrid cloud-fog computing architecture within a smart home environment consisting of 50 IoT devices. The performance is assessed using seven simulation tools: MATLAB, iFogSim, EdgeCloudSim, YAFS, FogNetSim++, PureEdgeSim, and LEAF. Simulation results demonstrate that distributing tasks between two fog nodes significantly reduces the execution time of high-priority tasks to approximately 0.0009 s. In contrast, MATLAB’s single-node approach achieves an execution time of 0.001 s but results in excessive CPU utilization, with Fog Node 1 reaching 90% usage and consuming up to 25,935 kWh. Task distribution between Fog Node 1 and Fog Node 2 achieves a more balanced load, with CPU utilization reduced to approximately 70% on one node and 20% on the other. This balanced allocation leads to a per-node energy reduction of around 20%. Cloud usage remains steady at 15% CPU utilization, indicating that offloading lower-priority tasks to the cloud has minimal impact on overall energy consumption. The redistribution of tasks also reduces memory usage, as the utilization of Fog Node 2 drops to approximately 10% when workloads are distributed. Findings highlight that achieving balanced resource allocation is critical for reducing latency, improving energy efficiency, and maintaining steady throughput across simulation tools. Comparative results indicate that iFogSim, EdgeCloudSim, YAFS, FogNetSim++, PureEdgeSim, and LEAF offer greater scalability and energy efficiency than MATLAB, particularly for large-scale deployments. The insights from this study can be extended to other IoT domains, such as healthcare and industrial automation, where latency optimization and resource efficiency remain paramount.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1570-7873
1572-9184
DOI:10.1007/s10723-025-09802-9