A resource allocation method for fog computing considering user mobility

With the growing use of mobile devices and delay-sensitive applications, the need to move computing power to the proximity of users increases compared to cloud computing. However, fog nodes have limited computing and storage capacity and, therefore, do not perform well under congestion of requests. One of the causes of congestion and loss of user requests is the mobility of users across fog node areas. The purpose of this paper is to provide an efficient method for resource allocation in fog computing considering user mobility with the aim of increasing the acceptance rate of users' delay-sensitive requests and reducing the response time of requests. The proposed method predicts congestion in the fog node based on the mobility of users and the condition of its queue and, accordingly, decides whether to process the requests in the fog or to send them to the cloud. The experimental results indicated that the proposed method involves a lower response time and a higher acceptance rate than other existing methods. The results of the experiments show that the response time of the proposed method is reduced by 73, 77 and 10 percent compared to the three FCFS, Cloud-only and Delay priority algorithms, respectively. In the proposed algorithm, the request acceptance rate has increased by 44, 25 and 5 percent compared to the above three algorithms, respectively.