Task scheduling in fog enabled Internet of Things for smart cities

Internet of Things (IoT) is an essential information infrastructure for smart cities in data sensing, collecting, merging, transmitting, or even reverse controlling. Whereas, the devices in IoT are fundamentally different in processing abilities. Combined with fog computing, some stronger nodes in IoT who have better computing capabilities can be taken as fog nodes to help other weak nodes in task processing, such as data analysis, computing, etc. Consequently, a new problem is raised that how to schedule the tasks among different fog nodes, so as to achieve minimum task makespan and communication cost. In this paper, we proposed an Adaptive Doublefitness Genetic Task Scheduling (ADGTS) algorithm. By the collaborative scheduling of tasks and fog resources, ADGTS can optimize the task makespan and communication cost simultaneously, as well as flexibly adapt to different importance of makespan and communication cost in practical applications. The simulation results show that the proposed algorithm has obviously better performance than traditional Min-Min algorithm in task makespan, and can balance the performance of task makespan and communication cost additionally.