Key points are not available for this paper at this time.
Challenging 5G requirements have been proposed to enable the provision of more satisfactory services to users. One of the most critical 5G enhancements is extremely low latency compared with that of 4G. Various 5G applications will be supported by Cloud computing services. However, Cloud services sometimes have a long communication distance because of the physical location of data centers. Fog computing architecture is being introduced in 5G cellular networks to shorten the latency arising from the location of data centers. The communication delay as well as the computing delay must be considered for latency-sensitive applications. A Fog server serving many users will probably have a longer computing delay than a Fog server with a lighter workload. To clarify the computing delay and communication delay in Fog architecture, we define a mathematical model of a Fog network and the important related parameters. We also analyze results from a model used to evaluate three different policies for selecting the target Fog server for each task. We conducted simulations to evaluate this model.
Intharawijitr et al. (Tue,) studied this question.