基于无服务器边缘计算下的服务负载调度算法  

Service load scheduling algorithm based on serverless edge computing

作　　者：高明[1] 陈国扬 Gao Ming;Chen Guoyang(School of Information&Electronic Engineering,Zhejiang Gongshang University,Hangzhou 310018,China)

机构地区：[1]浙江工商大学信息与电子工程学院,杭州310018

出　　处：《计算机应用研究》2024年第3期811-817,841,共8页Application Research of Computers

基　　金：国家自然科学基金资助项目(61871468);浙江省基础公益研究计划资助项目(LGG20F010015);浙江省新型网络标准与应用技术重点实验室基金资助项目(2013E10012)。

摘　　要：随着边缘计算的不断发展,其在资源管理配置方面逐渐出现相关问题,无服务器计算作为一种新的方式可以有效解决边缘计算的相关问题。然而,无服务器计算不具备在分布式边缘场景中高效处理请求所需服务负载调度的能力,针对这一问题,提出了一种基于无服务器边缘计算的服务负载调度算法(service load scheduling algorithm,SLSA)。SLSA的核心是通过隐式建模充分考虑了动态变化的节点状态、负载调度器放置等影响因素来优化整体时延,然后通过改进的平滑加权轮询调度(smooth weighted round robin,SWRR)算法进行服务调度。经仿真实验分析,SLSA在资源消耗上有着明显下降,同时在单城市场景与多城市场景下均有良好的性能表现,其中在单城市场景中相对于集中式轮询调度(round robin centralized,RRC)算法提升了43.01%,在多城市场景中提升了53.81%。实验结果表明,SLSA可以有效降低资源消耗率并提升性能。With the continuous development of edge computing,there are gradually related problems in resource management and configuration.As a new way,serverless computing can effectively solve the related problems of edge computing.However,serverless computing does not have the ability to efficiently process requests in distributed edge scenarios.To solve this problem,this paper proposed a service load scheduling algorithm(SLSA)based on serverless edge computing.The core of SLSA was to fully consider the dynamic changes of node status,load scheduler placement and other influencing factors through implicit modeling to optimize the overall delay,and then used the improved smooth weighted round robin(SWRR)scheduling algorithm for service scheduling.The simulation results show that SLSA has a significant reduction in resource consumption,and has good performance in both single-city scenarios and multi-city scenarios.In the single-city scenario,SLSA is 43.01%higher than the RRC algorithm.It improves 53.81%in multi-city scenarios.Experimental results show that SLSA can effectively reduce the resource consumption rate and improve the performance.

关 键 词：边缘计算 无服务器计算 负载调度 性能对比 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]