Finite Buffer Queueing System Performance Study with Multi Heterogeneous Fog Architecture

When it comes to new and developing applications that require high computation and low latency, fog computing is an attractive option to consider for implementation. The peripheral network devices are able to share idle resources and work together to complete fog computing tasks. It's up to the task publishers to decide how to distribute computer jobs that take into account transmission line quality and energy utilization. To urge these gadgets to engage in computational offloading, effective stimuli approaches are needed to stimulate them. Three heterogeneous servers with varying service rates are investigated using a finite constant buffer queue. Depending on a Poisson distribution with a mean and service times, the jobs are exponentially spread by distinct mean values There is a limit to how many jobs can be held in a queue while the servers are busy. It's assumed that Because of this, jobs are considered lost if they arrive after a buffer has reached capacity. If there is no client in queue, the next customer will be presumed to be in the first server if there is no customer in line. In the case of routers, this can be seen in their different speeds when serving packets on the network.