Fine-grained Task Scheduling Based on Priority for Heterogeneous Mobile Edge Computing

Le résumé fourni par la source

With the advance of novel applications, it is harder to deal with computation-intensive tasks due to the resource-constrained devices. Task offloading in Mobile Edge Computing (MEC) can effectively deal with this problem. Most of the existing studies view the task as a whole and do not consider the partition of task, which may result in the increasing delay of task processing. First, we construct a model of fine-grained task scheduling problem for heterogeneous MEC and consider the task as a Directed Acyclic Graph (DAG). Second, to optimize the average delay, we propose a priority-based heuristic task scheduling algorithm to handle dependencies of subtasks and achieve reasonable scheduling schemes between tasks and heterogeneous servers. Moreover, we design an idle time slot insertion strategy to realize the full utilization of MEC resources. Finally, the topology of DAGs is generated for simulation based on practical applications. Experimental result indicates that the proposed HFGO-CI could efficiently achieve the goal of average system delay reduction on the basis of reasonable decisions.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.