Delay-Aware Resource Management for Heterogeneous Service Collaboration in Mobile Edge Networks

Mobile Edge Computing (MEC) is introduced as a typical architecture of Mobile Edge Networks (MEN) characterized by low-latency and large computation capability, catering to emerging requirement of instant interaction for applications. Moreover, caching and offloading works with MEC are proved to efficiently decrease the delay of content delivery service and local computing service, respectively. However, past researches neglect the effect on resource management when multiple services take place in MEC network and what is the optimal division of resources for a network- centric revenue. In this paper, we propose a strategy of delay-aware resource management with the consideration of collaborated heterogeneous services in MEC network. We design an effective algorithm in pursuit of delay minimization with much lower time- complexity, and further explore the resource allocation scheme corresponding to different service revenues to address this issue. Numerical results show that the proposed strategy gets a maximal improvement of delay-aware evaluation with 72% than that of randomly offloading scheme. Simulation also reveals the distribution of the solution for resource allocation in terms of service revenues, which ensures the minimal delay of the network.