LAUSR

In the fog computing paradigm, fog nodes are placed on the network edge to meet end-user demands with low latency, providing the possibility of new applications. Although the role of the cloud remains unchanged, a new network infrastructure for fog nodes must be created. The design of such an infrastructure must consider user mobility, which causes variations in workload demand over time in different regions. Properly deciding on the location of fog nodes is important to reduce the costs associated with their deployment and maintenance. To meet these demands, this paper discusses the problem of locating fog nodes and proposes a solution which considers time-varying demands, with two classes of workload in terms of latency. The solution was modeled as a mixed-integer linear programming formulation with multiple criteria. An evaluation with real data showed that an improvement in end-user service can be obtained in conjunction with the minimization of the costs by deploying fewer servers in the infrastructure. Furthermore, results show that costs can be further reduced if a limited blocking of requests is tolerated.