基于机器学习的虚拟设备选择策略的设计
|
摘要
为了解决虚拟设备资源利用率低,能源消耗高等问题,设计了基于机器学习的虚拟设备决策算法.首先利用机器学习提出虚拟设备决策的优化算法,将机器学习与传统虚拟设备决策算法结合提出了基于机器学习的虚拟设备决策算法优化,解决了设备受资源限制的问题.然后利用机器学习技术中的DQN算法完成对虚拟设备决策算法的实现,通过系统自主学习,分析用户需求及软硬件环境,提出多种虚拟设备配置方案,最后经过系统对比选择出符合用户需求,同时资源消耗最优的虚拟设备配置方案.
In order to solve the problems of low resource utilization and high energy consumption of virtual equipment, a decision-making algorithm of virtual equipment based on machine learning is designed. Firstly, an optimization algorithm for virtual equipment decision-making is proposed based on machine learning, which combines machine learning with traditional virtual equipment decision-making algorithm and solves the problem of equipment resource constraints. Then, the DQN algorithm in machine learning technology is used to implement the decision-making algorithm of virtual equipment. Through the system autonomous learning, the user needs software and hardware environment are analyzed, and a variety of virtual equipment configuration schemes are proposed. Finally, through the system comparison, the virtual equipment configuration scheme which meets the user needs and consumes optimal resources is selected.
In order to solve the problems of low resource utilization and high energy consumption of virtual equipment, a decision-making algorithm of virtual equipment based on machine learning is designed. Firstly, an optimization algorithm for virtual equipment decision-making is proposed based on machine learning, which combines machine learning with traditional virtual equipment decision-making algorithm and solves the problem of equipment resource constraints. Then, the DQN algorithm in machine learning technology is used to implement the decision-making algorithm of virtual equipment. Through the system autonomous learning, the user needs software and hardware environment are analyzed, and a variety of virtual equipment configuration schemes are proposed. Finally, through the system comparison, the virtual equipment configuration scheme which meets the user needs and consumes optimal resources is selected.
引文
[1] 周志华.机器学习[M].北京:清华大学出版社,2016.
[2] 张茂红,宋艳,宫雪艳.虚拟化技术与云计算的完美结合[J].福建电脑,2015(10):79-80.
[3] 刘媛媛,高庆一,陈阳.虚拟计算环境下虚拟机资源负载均衡方法[J].计算机工程,2010,36(16):30-32.
[4] 朱宝珠,刘斌.考虑低能耗的云计算虚拟机部署方法[J].计算机工程与设计,2017,38(9):2319-2322+2328.
[5] 朱亚会,陈丹,庄毅.云数据中心资源利用率均衡的虚拟机调度算法[J].小型微型计算机系统,2017,38(2):232-237.
[6] 周晓彦,王珂,李凌燕.基于深度学习的目标检测算法综述[J].电子测量技术,2017,40(11):89-93.
[7] 黄雯坚.TensorFlow实战[M].北京:电子工业出版社,2017.
[8] 张立新,张栩之,吕海洋.基于云计算SME-FFD算法的概率有毒虚拟机资源配置[J].计算机工程,2016,42(5):23-29.
[2] 张茂红,宋艳,宫雪艳.虚拟化技术与云计算的完美结合[J].福建电脑,2015(10):79-80.
[3] 刘媛媛,高庆一,陈阳.虚拟计算环境下虚拟机资源负载均衡方法[J].计算机工程,2010,36(16):30-32.
[4] 朱宝珠,刘斌.考虑低能耗的云计算虚拟机部署方法[J].计算机工程与设计,2017,38(9):2319-2322+2328.
[5] 朱亚会,陈丹,庄毅.云数据中心资源利用率均衡的虚拟机调度算法[J].小型微型计算机系统,2017,38(2):232-237.
[6] 周晓彦,王珂,李凌燕.基于深度学习的目标检测算法综述[J].电子测量技术,2017,40(11):89-93.
[7] 黄雯坚.TensorFlow实战[M].北京:电子工业出版社,2017.
[8] 张立新,张栩之,吕海洋.基于云计算SME-FFD算法的概率有毒虚拟机资源配置[J].计算机工程,2016,42(5):23-29.