Jiale Zhao (ChongQing University, China); Mengxuan Dai (Jiangxi Normal University, China); Yunni Xia (Chongqing University, China); Yong Ma (Jiangxi Normal University, China); Meibin He (Department of Big Data and AI of Jiangxi Telecom, China); Kai Peng (Huaqiao University, China); Jianqi Li (Global Energy Interconnection Research Institute Co. Ltd, China); Fan Li (Sichuan University, China); Xiaodong Fu (Kunming University of Science and Technology, China)
Mobile Edge Computing (MEC) provides users with low-latency, highly responsive services by deploying Edge Servers (ESs) near applications. MEC allows any edge-hosted application or service to be migrated between different edge resource providers without being locked into a single provider. Nevertheless, due to its complexity and dynamics, the real edge computing environment is prone to errors and failures, reducing the reliability of edge service migration. This paper proposes a novel fault-tolerant method for redundant path service migration. The method utilizes sliding-window-based model and identifies a set of service migration paths, enabling the evaluation of the time-varying failure rate of ESs. The method combines resubmission and replication mechanisms and decides the edge service migration scheme by selecting multiple redundant migration paths. We also conduct extensive simulations and show that our proposed method outperforms traditional solutions in several metrics.