Evaluation of the Employment of UAVs as Fog Nodes

In fog computing, processing, network, and storage resources are placed close to end users to ensure low latency in comparison to the latency experienced when accessing the cloud. One limitation of this solution, however, is that fog nodes are usually fixed, whereas demands are variable over time at all locations, which may result in either under- or overprovisioning. This limitation incurs high CAPEX and OPEX costs to cope with user demands. One approach to address this problem is to employ mobile fog nodes dispatched to various locations to cope with the variability in resource demand. This article aims to evaluate the use of unmanned aerial vehicles (UAVs) equipped with processing elements as an alternative to fixed nodes in a fog infrastructure to cope with the variable workload in a metropolitan area. Although previous approaches considered UAVs as part of the network, they did not deploy multiple fog nodes mounted on battery-constrained UAVs. Specifically, we propose in this article a solution to the fog node location problem considering both fixed and mobile nodes to evaluate potential replacements of fixed servers with UAVs. Experimental evaluation of the problem using data generated by real mobile users shows that UAVs can indeed replace parts of the fixed fog infrastructure.