Optimization of task completion time in the Internet of Things using a modified particle swarm algorithm

In the Internet of Things (IoT) environment, especially in the medical sector, we are faced with a large volume of raw data and frequent requests that require fast response to requests and information. This challenge is especially complicated by the use of advanced technologies such as cloud computing, due to the instability of resources. To optimize resource management, tasks that can be transferred to the cloud are divided into groups using the K-means algorithm. Then, for each service, a virtual machine (VM) is created with the help of the Particle Swarm Optimization (PSO) algorithm to optimally migrate and transfer tasks between resources. To evaluate the quality of service, the task completion time parameter (Makespan) is considered and the CloudSim simulator and the PlanetLab standard dataset are used. The research results show that the proposed method was more successful in improving Makespan time than genetic algorithms, PABFD and PSO by 4.86%, 6.81% and 5.73%, respectively.