Computing resource allocation and task offloading in wireless edge network via genetic algorithm

Abstract

In recent years, a growing number of mobile devices are requesting computation-intensive tasks with low delay requirements. Cloud computing, which computes tasks at central servers remote from mobile devices, cannot bring access delay down to a certain level. In edge computing, on the other hand, mobile devices offload their tasks to servers installed on edge nodes that are closer to the mobile devices. This enables the user tasks to be completed within the delay requirement. In this scenario, determining how to allocate computing resources on edge nodes is critical to the access delay of users. In addition, when edge servers are installed, the total service delay of devices offloading tasks can be minimized by properly determining which servers to offload them. However, existing researches on server placement are restricted to the proposal of two-phase optimization of each issue with heuristic algorithms. In this paper, we study the joint design of computing resource allocation and task offloading in wireless edge network. Based on the genetic algorithm, we designed a new algorithm for a given problem and used it to find computing resource allocation and task offload scheme that minimizes the average service delay while satisfying the delay requirement of individual users. The simulation results show that the proposed algorithm performs very well.