On-Demand Service Deployment Strategies for Fog-as-a-Service Scenarios

Service deployment at the network edge is a promising area that has been studied recently in the literature. In this work we have investigated a Fog-as-a-Service scenario, where multiple Server Fog Nodes (SFNs) can serve multiple Client Fog Nodes (CFNs) by exploiting different service deployment models, i.e., SaaS, PaaS, and IaaS, in a flexible way. The system has been modeled as a Size-Constrained Weighted Set Cover Problem aiming at maximizing the amount of satisfied CFNs exploiting a heterogeneous service deployment architecture, while minimizing the service completion time in a computation offloading scenario. In the simulation results section, we analyze the performance of different methods in terms of percentage of CFNs' offloading requests satisfaction and offloading delay.