• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (11): 2018-2024.

• 论文 • 上一篇    下一篇

面向高校教学云服务的虚拟机动态调度方法

王菁,王岗,高晶,李寒,马倩   

  1. (1.北方工业大学云计算研究中心,北京 100144;
    2.北方工业大学大规模流数据集成与分析技术北京市重点实验室,北京 100144)
  • 收稿日期:2015-08-20 修回日期:2015-09-30 出版日期:2015-11-25 发布日期:2015-11-25
  • 基金资助:

    北京市属高等学校创新团队建设与教师职业发展计划基金资助项目(IDHT20130502);北京市教育委员会科技计划重点项目(KZ201310009009)

A dynamic scheduling method of virtual machines
for campus cloud teaching services 

WANG Jing,WANG Gang,GAO Jing,LI Han,MA Qian   

  1. (1.Cloud Computing Research Center,North China University of Technology,Beijing 100144;
    2.Beijing Key Laboratory on Integration and Analysis of Largescale Stream Data,
    North China University of Technology,Beijing 100144,China)
  • Received:2015-08-20 Revised:2015-09-30 Online:2015-11-25 Published:2015-11-25

摘要:

随着教学信息化的不断深化,高校云平台越来越普及,但是实际应用中资源利用率仍然较低,核心问题在于当前的虚拟机调度机制未考虑高校教学应用的特征,从而导致负载不均和资源浪费。为了解决这一问题,从高校教学应用需求出发,提出了一种虚拟机动态调度算法(CRS),定义了课程虚拟机模型和物理机负载模型,并实现了基于OpenStack开源云平台的可对虚拟机进行动态调度的校园云平台。实验表明,提出的虚拟机动态调度方法达到了降低能耗及实现负载均衡的目标。

关键词: 校园云, 动态调度, 节能, 负载均衡

Abstract:

With the continuous deepening of teaching informatization, the campus cloud platform is becoming increasingly popular, but the resource utilization in practical applications is still low. The key problem is that the current scheduling mechanism of virtual machines does not take into account the characteristics of campus teaching applications, resulting in a waste of resources and load unbalancing. To tackle this problem, according to the requirements of teaching applications in universities, we put forward a dynamic scheduling algorithm of course virtual machines(CRS), define the course requirement model and the physical machine load model, and implement a campus cloud platform based on the OpenStack. Experimental results show that the platform achieves energy consumption reduction and load balancing.

Key words: campus cloud;dynamic scheduling;energy saving;load balancing