Resource Management Across Edge Server in Mobile Edge Computing

Mobile Edge Computing (MEC) has the potential to provide computational resources to edge devices. The Edge device (Ed) does not have enough computational resources, so it offloads its tasks to nearby Mobile Edge Servers (MES), while sometimes offloading tasks to MES causes MES to overload. The overload state of MES is when the data occupies all the CPU cores and memory of MES, so MES rejects requests for data offloading from other Eds. In this research, we have considered a multi-server, multi-device scenario, for which we have proposed an algorithm called CPU Memory Mobility Resource Allocation (CMMRA) that does not reject Ed from providing computational resources and encourages them to migrate the task to another MES in case the current MES is overloaded. The algorithm increases the service rate of a communication area, the rejection rate of the devices is also reduced, the total processing time of MES in this research is less as compared to others, and the migration in our proposed algorithm is greater than others. As we consider the mobility of devices and many devices are requesting computational resources, if the current MES is busy, it migrates the task to an underloaded MES for computation.