User-participatory fog computing architecture and its management schemes for improving feasibility

Abstract

The evolution of computing and networking technologies has opened the era of cloud computing, and the advent of the Internet of Things (IoT) paradigm has been questioning its limitations. Owing to advances in computer networks, cloud computing is improving, and the most promising technology is fog computing. Although fog computing is recognized as the most appropriate computing model for the IoT, it has not yet been widely used, and the major reasons are as follows. The replacement of the firmware and hardware of network equipment is inevitable; however, the operator in charge of carrying out this expensive task is unclear, and even if the operator is selected, the reason may be not rational. In addition, although fog computing is based on collaboration between several infrastructure operators and service providers, it is not clear who operates and manages the infrastructure. Furthermore, there is still a resource allocation problem for a fog service instance. In this paper, we propose a user participatory fog computing architecture and its management schemes to address the above problems related to its feasibility. In the proposed architecture, fog service instance placement optimization is performed based on service usage of participating users, which is formulated into a mixed-integer non-linear programming problem and then linearized. The proposed architecture and the fog service placement method are evaluated based on simulation, taking into account actual parameters of the IoT services and devices.