Modeling and Analysis of Scheduling in Edge Server Using Network Calculus

With the explosive growth of data traffic flow, mobile edge computing (MEC) has been suggested as an advancement of conventional cloud computing. However, as the spread of emerging applications and the requirement for quality of service (QoS) increases, the performance of the edge network has been an important issue. In order to provide better service, in this paper, we focus on the queuing problem in a single edge server and propose a scheduling algorithm in which tasks are placed in the different queues and the higher priority is given to the delay-sensitive task. Besides, we derive the theoretical formulation for the service curve under this algorithm with network calculus, and then derive the delay bounds. Finally, the numerical results are presented to show that the delay is significant reduced in our proposed algorithm compared with the existing algorithms.