Task Offloading Strategy and Pricing Scheme in Fog-Enabled Networks

Abstract

This paper investigates the price-based task offloading in fog-enabled networks. Specifically, we consider a fog-enabled network with one task node (TN) and multiple helper nodes (HNs) where the TN and each HN are assumed to be M/M/1 queuing systems. Considering that the tasks are not processed for free at the HNs, we address the optimal task offloading strategy for the TN and optimal pricing scheme for the HNs, respectively. In particular, assuming the equal quality of service (EQoS) at the HNs, a low complexity algorithm is proposed to obtain the optimal task offloading strategy for the TN when the pricing scheme for the HNs is given. Meanwhile, by taking into account the optimal task offloading strategy for the TN, we derive a necessary condition that the optimal pricing scheme for the HNs must satisfy. Furthermore, a low complexity algorithm is proposed to obtain one near optimal pricing scheme for the HNs. Numerical results demonstrate the advantage of the proposed algorithms.