Placement of the conscious energy virtual machine based on the PAPSO particle swarm optimization method

One of the most important issues of cloud computing is the issue of load balancing and reducing the energy consumption of physical cloud hosts, this is important by migrating virtual machines from overloaded hosts to hosts with minimal energy consumption. Virtualization is one of the effective and useful ideas in the direction of increasing productivity, which provides the possibility of combining and integrating different resources and prepares them to perform different tasks, so that all users feel access to their own environments. In order to optimize the overall quality of cloud services, the multi-objective function including load balancing, maximizing utilization, and minimizing energy consumption is used. To optimize this function, we proposed a meta-heuristic particle swarm optimization (PAPSO) scheduling algorithm. This algorithm is for dynamic task scheduling, workflow scheduling and load balancing. In the proposed method, the fair distribution of workload in cloud data centers is realized along with the reduction of energy consumption. Since task scheduling on cloud data center processors is an NP problem with multiple conflicting, a two-phase method is proposed using the aforementioned algorithm. The simulation results show that the proposed PAPSO algorithm provides better results than similar methods and other meta-heuristic algorithms. CPU efficiency in physical machines in proposed method performs better on ۲۰ physical machines than on ۴۰ and ۶۰ machines. The processor efficiency in physical machines in the proposed method shows a greater difference in ۵۰۰ virtual machines compared to MDPSO and EALBPSO methods. Energy consumption in the proposed method by increasing the number of virtual machines compared to the timing of the processors in terms of the standard deviation of the proposed method in the number of ۳۰۰ processors has performed better than other models.