Two Hybrid Heuristic Algorithms for Optimizing Data Processing Schedules in Distributed Edge Computing

Abstract

In recent years, edge computing has emerged as a promising solution in the field of network computing. This architecture ensures the availability of distributed computing resources located closer to end-users and IoT devices. However, resource scheduling remains a significant challenge in edge computing, requiring effective strategies to optimize resource utilization and ensure efficient task allocation. In this paper, we propose two hybrid approaches that combine the Nawaz-Enscore-Ham (NEH) algorithm with local search and Greedy Random Adaptive Search Procedure (GRASP) algorithm with local search for modeling and solving data traffic in distributed edge computing environments (DPSDEC). Through extensive evaluations, we consistently observe that the NEH algorithm outperforms GRASP, delivering minimized makespan and generating efficient schedules. Moreover, the NEH algorithm performs very well in less complex situations and maintains this advantage even in larger and more complex problems.