A Queueing Game Approach for Fog Computing with Strategic Computing Speed Control
Changyan Yi, Jun Cai · 2019
In this paper, a novel queueing game framework for computational workload assignment and strategic computing speed control in fog computing is proposed. Unlike most existing studies in the literature, our work jointly addresses two practical issues related to fog computing, i.e., i) computation tasks offloaded by mobile users are generated dynamically over the time; and ii) each third-party fog node can strategically allocate its computing resource (or control its computing speed) for balancing the tradeoff between the reward gained from executing offloaded computation tasks and the cost incurred by the dissatisfied service to its own subscribed tasks. To describe the long-term performance of the computation task distribution/assignment and inherent strategic interactions among fog nodes, a noncooperative game upon a queueing model is formulated. Based on this, an adaptive algorithm is designed to jointly determine the optimal task distribution and the equilibrium computing speed of each fog node. Theoretical analyses and simulation results examine the performance of the proposed approach and demonstrate its superiority over counterparts.