Affiliation:
1. Computer Science Department, Sapienza University of Rome, 00185 Rome, Italy
Abstract
The Internet of Things (IoT) uptake brought a paradigm shift in application deployment. Indeed, IoT applications are not centralized in cloud data centers, but the computation and storage are moved close to the consumers, creating a computing continuum between the edge of the network and the cloud. This paradigm shift is called fog computing, a concept introduced by Cisco in 2012. Scheduling applications in this decentralized, heterogeneous, and resource-constrained environment is challenging. The task scheduling problem in fog computing has been widely explored and addressed using many approaches, from traditional operational research to heuristics and machine learning. This paper aims to analyze the literature on task scheduling in fog computing published in the last five years to classify the criteria used for decision-making and the technique used to solve the task scheduling problem. We propose a taxonomy of task scheduling algorithms, and we identify the research gaps and challenges.
Subject
Computer Networks and Communications
Reference75 articles.
1. Fog Computing, Applications, Security and Challenges, Review;Rahman;Int. J. Eng. Technol.,2018
2. Performance Evaluation Metrics for Cloud, Fog and Edge Computing: A Review, Taxonomy, Benchmarks and Standards for Future Research;Aslanpour;Internet Things,2020
3. Dynamic Scheduling for Stochastic Edge-Cloud Computing Environments Using A3C Learning and Residual Recurrent Neural Networks;Tuli;IEEE Trans. Mob. Comput.,2022
4. Abdel-Kader, R., El Sayad, N., and Rizk, R. (2021). Efficient energy and completion time for dependent task computation offloading algorithm in industry 4.0. PLoS ONE, 16.
5. Multi-agent deep reinforcement learning for task offloading in group distributed manufacturing systems;Xiong;Eng. Appl. Artif. Intell.,2023