多接入边缘计算中相关性任务的联合调度算法

Joint scheduling algorithm for correlative tasks in multi-access edge computing

多接入边缘计算已经成为资源密集型应用程序的有前途的计算范式。不过,先前大部分研究工作没有考虑到任务的相关性,这可能导致不可行的调度决策。考虑应用程序上有些任务必须要在本地完成,研究了相关性任务在本地和边缘侧的联合调度方法,并考虑了多接入边缘计算卸载场景下的另一个不可忽视的能耗问题。将问题形式化为在满足应用程序的完成截止时间约束的条件下最小化系统中的能耗,并提出联合调度(JS)算法解决该问题。最后通过仿真实验分析JS算法在应用程序的完成率和系统能耗两方面的性能。仿真结果表明,JS算法在应用程序的完成率上优于其他对比算法并且至少可以节省43%的系统能耗。

Multi-access edge computing(MEC)has emerged as a promising computing paradigm for resource-intensive applications.However,most of the previous research work has not considered correlative tasks,which may lead to infeasible scheduling decisions.Considering that some tasks on the application must be completed locally and another non-negligible energy consumption problem in the multi-access edge computing offloading scenario,the joint scheduling algorithm of correlative tasks on the local and edge sides was studied.The problem was formalized as minimizing the energy consumption in the system while satisfying the application’s completion deadline constraints,and the joint scheduling(JS)algorithm was proposed to solve the problem.Finally,the performance of the JS algorithm in the application completion rate and system energy consumption were analyzed through simulation experiments.The simulation results show that the JS algorithm is superior to other comparison algorithms in the application completion rate and can save at least 43%of the system energy consumption.