Distributed Multi-agent Cooperative Resource Sharing Algorithm in Fog Networks
Yunjun Zheng, Weiwei Xia, Long Jiang, Feng Yan, Lianfeng Shen · 2020
Compared with traditional cloud computing technology, fog computing provides lower latency services in the next generation mobile networks. However, the imbalance of workloads and computing resources among Fog Nodes (FNs) restricts the further promotion of the network service quality. In this paper, considering the resource sharing of FNs by cooperation where FNs are as agents, we propose a distributed multi-agent cooperative resource sharing (DMCRS) algorithm to minimize the average service latency of the fog network. Firstly, the multi-agent resource sharing problem of the fog network is formulated as a convex optimization problem and the convexity is proved. Then the coalitional graph game framework is applied to achieve the cooperation among agents and the coalition graph Nash equilibrium is proved. To reduce the complexity, the global optimization problem is decomposed into distributed local optimization sub-problems in the DMCRS algorithm and these sub-problems are transformed into root-finding problems of polynomial equations. Simulation results demonstrate that the DMCRS algorithm can balance the workloads of different FNs, reduce the average service latency remarkably and the convergence result of the algorithm is close to the global optimal solution. In addition, the time complexity of the DMCRS algorithm is significantly lower than existing algorithms.