.png){kind=icon}
Return
Multi-Objective Optimization for Task Offloading and Resource Allocation via End-Edge-Cloud Collaboration
.png){kind=icon}
PRE
AI
DOI：10.23919/cje.2025.00.179.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Compared to traditional edge computing systems, edge-cloud hybrid systems are increasingly effective in meeting the demands for low-latency and efficient task offloading. However, conventional edge-cloud computing often focuses solely on optimizing energy consumption, latency, or a combination of both, frequently over-looking the need for load balancing in task allocation among end devices, edge servers, and cloud servers. This paper explores an optimization framework that integrates the virtual machine (VM) architecture with an end-edge-cloud collaboration network (VM-ECN) for task offloading. We formulate a joint optimization problem that takes into account latency, energy consumption, and load balancing, facilitating task offloading through collaboration among end devices, edge servers, and cloud servers. To address it, we propose an improved multi-objective grey wolf optimization algorithm that incorporates tent mapping and a multi-strategy mechanism, referred to as tent mapping-based multi-objective grey wolf optimizer (TMOGWO). We conduct a comprehensive evaluation of the multi-objective optimization problem within the VM-ECN system using the analytic hierarchy process and perform a series of experimental simulations with various parameters. The results show that our proposed TMOGWO achieves better convergence of composite indicator values across four distinct task scenarios compared to non-dominated sorting genetic algorithm II (NSGAII), multi-objective particle swarm optimization (MOPSO), multi-objective grey wolf optimizer (MOGWO). Furthermore, TMOGWO demonstrates superior performance compared to NSGAII, MOPSO, and MOGWO in load-balancing metrics, while also achieving enhanced centralized distribution and showing increased stability with respect to time delay and energy consumption.
Keywords:
Algorithms
Clouds
Internet of Things
Optimization
Servers
Cloud computing
Energy consumption
Equations
Timing
Virtual machines
Task offloading
Multi-objective optimization
End-edge-cloud collaboration
Virtual machine
Mobile edge computing
Multi-objective grey wolf optimization algorithm
Analytic hierarchy process
Journal
C
IF:
3
Papers:
97
Citations:
1.7K
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveShareSaveShareSaveShareSaveShareSave
View more