Dependent Task Offloading in Edge Computing Using GNN and Deep Reinforcement Learning
IEEE Internet of Things Journal · 15.06.2024 · 11 (12) · 21632-21646
DOI: 10.1109/jiot.2024.3374969 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 15.06.2024. Verfasst von Zequn Cao und 5 weiteren Autor:innen. Die Arbeit wurde bisher 76-mal zitiert (Crossref) und nennt 49 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Joint Funds for Railway Fundamental Research of National Natural Science Foundation of China, Special Fund of National Key Laboratory of Ni&Co Associated Minerals Resources Development and Comprehensive Utilization.
Abstract (Originalsprache)
Task offloading is a widely used technology in edge computing (EC), which declines the makespan of user task with the aid of resourceful edge servers. How to solve the competition for computation and communication resources among tasks is a fundamental issue in task offloading. Besides, real-life user tasks often comprise multiple interdependent subtasks. Dependencies among subtasks significantly raises the complexity of task offloading, and makes it difficult to propose generalized approaches for scenarios of different sizes. In this article, we study the dependent task offloading (DTO) problem within both single-user single-edge and multiuser multiedge scenario. First, we use directed acyclic graph (DAG) to model dependent task, where nodes and directed edges represent the subtasks and their interdependencies, respectively. Then, we propose a task scheduling method based on graph attention network (GAT) and deep reinforcement learning (DRL) to minimize the makespan of user tasks. More specifically, our method introduces a multidiscrete action DRL scheduler that simultaneously determines which subtask to consider and whether it should be offloaded at each step, and employs GAT to encode the graph-based state representation. To stabilize and speed up DRL scheduler training, we pretrain GAT encoder with unsupervised learning. Extensive experiments demonstrate that our proposed approach can be applied to various environments and outperforms prior methods.
Förderung
- National Natural Science Foundation of China
- Joint Funds for Railway Fundamental Research of National Natural Science Foundation of China
- Special Fund of National Key Laboratory of Ni&Co Associated Minerals Resources Development and Comprehensive Utilization
- Key Project of Shenzhen City Special Fund for Fundamental Research
- National Natural Science Foundation of Hunan Province
Zitieren
APA
Cao, Z., Deng, X., Yue, S., Jiang, P., Ren, J., & Gui, J. (2024). Dependent Task Offloading in Edge Computing Using GNN and Deep Reinforcement Learning. IEEE Internet of Things Journal, 11(12), 21632-21646. https://doi.org/10.1109/jiot.2024.3374969
Vancouver
Cao Z, Deng X, Yue S, Jiang P, Ren J, Gui J. Dependent Task Offloading in Edge Computing Using GNN and Deep Reinforcement Learning. IEEE Internet of Things Journal. 2024;11(12):21632-21646. doi:10.1109/jiot.2024.3374969
BibTeX
@article{Cao2024,
author = {Cao, Zequn and Deng, Xiaoheng and Yue, Sheng and Jiang, Ping and Ren, Ju and Gui, Jinsong},
title = {Dependent Task Offloading in Edge Computing Using GNN and Deep Reinforcement Learning},
journal = {IEEE Internet of Things Journal},
year = {2024},
volume = {11},
number = {12},
pages = {21632--21646},
doi = {10.1109/jiot.2024.3374969}
}
EUVIMED