Fog Computing In Multi-Tier Data Center Networks: A Hierarchical Game Approach

Huaqing Zhang,Yong Xiao,Shengrong Bu,Dusit Niyato,Richard Yu,Zhu Han
DOI: https://doi.org/10.1109/ICC.2016.7511146
2016-01-01
Abstract:With the increasing popularity of data services and applications, data center networks have been introduced to serve users in a centralized fashion. Furthermore, the flexibility of the data service subscribers' (DSSs') requirements motivates data center virtualization so as to optimize the resource allocation among all DSSs. However, as the massive data centers are usually far away from the DSSs, the quality of services are severely affected for delay-sensitive DSSs. Accordingly, fog computing is considered to solve the problem, where some virtualized edge data centers acting as fog nodes (FNs) are added in the network and help massive data center operators (MDCOs) serve DSSs. In this paper, we analyze the resource management problem in the multi-FN multi-MDCO, and multi-DSS networks. We model the network architecture with 3-layer model, where the FNs are in the upper layer, MDCOs in the middle layer, the DSSs in the bottom layer. The FNs first share computing resource with the MDCOs, and thus the MDCOs are able to serve their DSSs with low delay. Based on the model, we propose a hierarchical game, where the interaction between FNSs and MDCOs is regarded as a multi-leader multi-follower Stackelberg game, and the interactions between MDCOs and DSSs are regarded as the single-leader single-follower Stackelberg games. By making decisions distributively, all FNs, MDCOs, and DSSs receive high utilities. Simulation results show the correctness of the analysis and the performance improvement of the proposed strategies in the fog computing networks.
What problem does this paper attempt to address?