摘  要： 如何在提高云计算服务资源调度效率的同时尽量降低工作能耗，成为当前必须解决的问题。因此，提出一种基于遗传算法的云计算资源调度方法，适用于高校师资培训资源管理。首先，搭建基于云服务的高校教师师资培训系统模式;其次，利用具有较强NP问题解决能力的遗传算法设计大规模集群的资源调度方法，并采用粒子群算法对其全局寻优能力进行改进;最后，设计了由质量和成本构成的双指标约束适应度函数。CloudSim平台的Hadoop实验仿真结果显示，提出的遗传云计算资源调度方法在约束条件下获得了最佳的调度结果。

关键词： 云计算; 遗传算法; Hadoop; 资源调度; 平衡性; 调度策略

中图分类号： TN02?34; TP393                     文献标识码： A                     文章编号： 1004?373X（2020）01?0157?04

Management of college teacher training based on genetic algorithm

in cloud computing of Hadoop platform

XING Lili

Abstract： How to improve the efficiency of cloud computing service resource scheduling while minimizing the energy consumption of the work as far as possible has become an issue that must be solved at present. Therefore， a cloud computing resource scheduling method based on genetic algorithm is proposed， which is suitable for the management of college teacher training resources. First of all， a cloud service based teacher training system model for college teachers is established. Secondly， a large?scale cluster resource scheduling method is designed by genetic algorithm with powerful NP problem solving ability， and its global optimization ability is improved by particle swarm optimization. Finally， a dual?index constrained fitness function consisting of quality and cost is designed. The Hadoop experimental simulation results of CloudSim platform show that the proposed genetic cloud computing resource scheduling method obtains the best scheduling results under the constraint conditions.

Keywords： cloud computing; genetic algorithm; Hadoop; resource scheduling; balance; scheduling strategy

0  引  言

自2007年以来，云计算已经逐渐成为各个行业领域争相构建的战略方向。许多基于云计算的产品和服务不断被推出，计算机行业的规模和涉及的领域也随之不断拓张。云计算辅助教学（Cloud Computing Aided Instruction，CCAI）已经成为了广大高校和教师搭建现代化、信息化和个性化教学的新手段，从技术上有效提高了教学的质量[1?2]。CCAI模式的特点十分有利于高校师资培训的信息化管理，降低了资金投入和维护成本，提高了网络安全性，有助于构建个性化教学环境。实际推广应用结构显示通过CCAI模式可以有效解决当前高校师资培训面临的空间限制和资源匮乏问题[3]。

近期基于CCAI模式的教学资源管理得到了较多的关注。例如，文献[4]提出了基于云计算的高校教育资源共享服务平台构建研究，实现了按需为教师及学生提供个性化数字资源。……