ABSTRACT
The most critical action in the Internet of things (IoT), particularly in the cloud computing (CC) ecosystem, is job/task scheduling. The rule of the task scheduling (TS) means that how to arrange/schedule the jobs over the given virtual machines (VMs) by decreasing the Makespan measured (Mm) values and the cost need. Various scholars propose several scheduling algorithms for solving this problem of scheduling the tasks in cloud computing ecosystems. In this chapter, a task scheduling method is proposed using a multi-objective design model and the Gray Wolf Optimizer (GWO), called TS-GWO. Firstly, the multi-objective criterion measures the fitness function by determining the cost value of the CPU ratio and memory size (Me). The fitness function is determined by calculating the Makespan value and demand value. The proposed method (TS-GWO) can expertly arrange the given jobs to the offered VMs while keeping the smallest Makespan value and cost value. Conclusively, the performance of the proposed method (TS-GWO) is investigated and matched with the basic GWO for the testing measures: Makespan ratio and cost value. From the empirical results, we concluded that the introduced scheduling method (TS-GWO) schedule the given tasks to the offered VMs effectively by producing the smallest Makespan values and the smallest average cost values.
