A parallel multi-objective imperialist competitive algorithm to solve the load offloading problem in mobile cloud computing

Cloud computing is a modern architecture for performing complex and immense processes. It consists of configurable computational resource sets that communicate with each other through communication networks. With the advent of the cloud computing architecture and increasing its applications for mobile devices, the growth rate of mobile data has proliferated exponentially. Consequently, processing the tasks of mobile users has become difficult due to the limitations of these devices, such as low computing power and low capacity. Therefore, the idea of mobile cloud computing (MCC) for mobile devices using cloud-based storage and computing resources was introduced. In MCC, processing information is transferred from the user's mobile devices to the cloud servers. This process is known as the tasks offloading and scheduling of mobile users. In this case, the task execution time, CPU power consumption, network bandwidth, and task allocation time must be specified. Due to many tasks and different resources, the process of task offloading and scheduling is considered a challenging subject in the field of MCC. Therefore, in this paper, a multi-objective parallel imperialist competitive algorithm (MPICA) is proposed. The main objective of this parallel algorithm is to reduce the algorithm's execution time for searching the problem space, reducing processing time, reducing energy consumption, and improving load balance. The simulation results of the proposed algorithm represent that the parallelization of the imperialist competitive algorithm (ICA) has a significant effect on reducing the execution time of the algorithm. In general, the proposed algorithm performs better than the state-of-the-art algorithms based on the proposed criteria.