Energy and task completion time trade-off for task offloading in fog-enabled IoT networks

• Published in: Pervasive and mobile computing Vol. 74; p. 101395
• Main Authors: Shahryari, Om-Kolsoom, Pedram, Hossein, Khajehvand, Vahid, TakhtFooladi, Mehdi Dehghan
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.07.2021
• Subjects: Fog computing; Genetic algorithm; Internet of Things; Particle swarm optimization; Resource allocation; Task offloading; Fog computing; Internet of Things; Genetic algorithm; Particle swarm optimization; Task offloading; Resource allocation
• ISSN: 1574-1192; 1873-1589
• DOI: 10.1016/j.pmcj.2021.101395

In order to improve the quality of experience in executing computation-intensive tasks of real-time IoT applications in a fog-enabled IoT network, resource-constrained IoT devices can offload the tasks to resource-rich nearby fog nodes. It causes a reduction in energy consumption compared with local processing, although it extends task completion time due to communication latency. In this paper, we propose a task offloading scheme that optimizes task offloading decision, fog node selection, and computation resource allocation, investigating the trade-off between task completion time and energy consumption. Weighting coefficients of time and energy consumption are determined based on specific demands of the user and residual energy of devices’ battery. Accordingly, we formulate the task offloading problem as a mixed-integer nonlinear program (MINLP), which is NP-hard. A sub-optimal algorithm based on the hybrid of genetic algorithm and particle swarm optimization is designed to solve the formulated problem. Extensive simulations prove the convergence of the proposed algorithm and its superior performance in comparison with baseline schemes. •We propose a task offloading algorithm for fog-enabled IoT networks.•We intend to minimize task completion time and energy consumption of IoT devices.•We employ a combination of GA and PSO to find a near-optimal solution for formulated MINLP problem.•We investigate the trade-off between task completion time and energy consumption.•We take into consideration the residual energy of the IoT device and QoS requirement of tasks.