A Real-Time Task Offloading Strategy Based on Double Auction for Optimal Resource Allocation in Edge Computing

Abstract

Task offloading in edge computing becomes an effective method to extend the computation ability of user equipments (UEs), via migrating computation-intensive applications from UEs to edge servers. However, not only locality-aware resource allocation for UEs and various edge computing services providers (ESPs) but also network economics for profit-driven ESPs and UEs is still a big challenge in task offloading. In this paper, we propose an edge computing resource allocation model based on the continuous-cycle double auction mechanism (RABDA). Considering the emergency of task offloaded, we also propose real-time offloading strategy (RTOS) to ensure tasks are processed efficiently. We use genetic algorithm to determine the winner ESPs which are responsible for providing computational resources to UEs, and verify the performance of our algorithm by contrast experiment. The simulation results show that our algorithm can improve satisfaction between UEs and ESPs, and it has higher resource utilization than the existing algorithm.