The Optimal Strategy To Deploy Virtual Network Function In Mobile Edge Computing

Network Function Virtualization(NFV)and Mobile Edge Computing(MEC)are the two core technologies in the fifth generation mobile communication network technology,therefore the service function chain allocation(SFC-A)in MEC plays an important role in enhancing the quality of experience(QoE)of the end users.Different from the SFC-A in the traditional cloud computing,SFC-A in MEC is more challenging due to the limited computing resources at edge servers in MEC.In this paper,we investigated the SFC-A problem in MEC with limited resources,and formulate the problem as an integer linear program(ILP)problem with the objective of minimizing both the transmission latency and processing latency.Since the SFC-A problem is an NP-hard problem,it is impossible to find the optimal solution in polynomial time.Therefore,this paper proposes the GA-SFC-A,which is a Genetic Algorithm-based method to solve this problem.Meanwhile,a series of experiments are set to evaluate the performance of this algorithm.Furthermore,to overcome the inherent shortcomings of heuristic solutions,this paper proposes an algorithm,DRL-SFC-A,which leverages deep reinforcement learning to make the optimal decision by taking advantage of its trial-and-error mechanism,reward mechanism and exploration-exploitation ability.This algorithm can avoid the result falls into the local optimal solution as much as possible,and also improves the performance of the algorithm.A series of simulations verified that the proposed algorithm is effective in reducing the average latency of SFC requests,as well as to provide QoE guarantee.