Reinforcement learning-based framework for optimizing task allocation and service migration in edge-cloud systems

View at publisher Summarise with VIRAT
Authors: Weiwei Guo, Feng Liu, Yingjie Song, Xiaomin Zhao
Cited by: 0
References: 9

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Addressing issues such as unreasonable resource allocation, limited computing power of edge devices, data privacy leakage, and difficulties in meeting low-latency and low-power consumption requirements in edge cloud computing models, a dynamic task allocation and service migration model for edge cloud systems is proposed. This model constructs a three-tier architecture consisting of resource allocation, data management, and intelligent collaboration. It employs a Markov decision process optimization algorithm to solve resource allocation challenges and establishes an experimental environment involving multiple edge devices and cloud servers. Experiments demonstrate that the model dynamically adjusts task priorities through three-dimensional evaluation, significantly reducing task completion time and enhancing resource utilization. Additionally, it utilizes a decision tree to optimize migration decisions, minimizing ineffective migrations, effectively compensating for the deficiencies of traditional scheduling models.

Fields of Study & Subjects

IoT and Edge/Fog Computing Cloud Computing and Resource Management Energy Efficiency in Computing

Publication details

Journal / source
International Conference on Optics, Computer Applications, and Materials Science (CMSD)
Pages
57–57
Publication date
1 April 2026
Language
EN