计算机应用 ›› 2015, Vol. 35 ›› Issue (12): 3530-3535.DOI: 10.11772/j.issn.1001-9081.2015.12.3530

• 计算机软件技术 • 上一篇    下一篇

基于虚拟机部署策略的云平台容错即服务方法

刘晓霞, 刘靖   

  1. 内蒙古大学计算机学院, 呼和浩特 010021
  • 收稿日期:2015-06-18 修回日期:2015-08-30 出版日期:2015-12-10 发布日期:2015-12-10
  • 通讯作者: 刘靖(1981-),男,内蒙古呼和浩特人,副教授,博士,CCF会员,主要研究方向:云计算、软件可靠性工程
  • 作者简介:刘晓霞(1990-),女,内蒙古呼和浩特人,硕士研究生,主要研究方向:云计算、软件容错。
  • 基金资助:
    国家自然科学基金资助项目(61262017);内蒙古自然科学基金资助项目(2015MS0611)。

Fault tolerance as a service method in cloud platform based on virtual machine deployment policy

LIU Xiaoxia, LIU Jing   

  1. College of Computer Science, Inner Mongolia University, Hohhot Nei Mongol 010021, China
  • Received:2015-06-18 Revised:2015-08-30 Online:2015-12-10 Published:2015-12-10

摘要: 针对如何充分利用云基础架构层资源,满足上层云应用系统租户对应用系统容错的需求多样性和高可靠性要求的问题,提出一种面向租户和云服务提供商的、基于虚拟机部署策略的云平台容错即服务方法。该方法根据租户的特定容错需求适配适合的容错方法及容错级别,据此计算并最优化云服务提供商的收益和资源使用量,在此基础上对提供容错服务的虚拟机进行优化部署,充分利用底层虚拟机资源为租户的云应用系统提供更为可靠的容错服务。实验结果表明,所提方法能够在保障云服务提供商收益的基础上,为多租户云应用系统实现更灵活且可靠性更高的容错服务。

关键词: 云应用系统, 虚拟机部署, 容错即服务, 高可靠性

Abstract: Concerning the problem that how to make full use of the resources in cloud infrastructure to satisfy various and high reliable fault tolerant requirements of cloud application systems for cloud application tenants, a cloud application tenant and service provider oriented fault tolerance as a service method in cloud platform was proposed based on virtual machine deployment policy. According to specific fault tolerant requirements from cloud application tenants, suitable fault tolerance methods with corresponding fault tolerant levels were adopted. Then, the revenue and resource usage of service provider were computed and optimized. Based on the analysis, virtual machines which providing fault tolerant services were well deployed, which could make full use of the resources in virtual machine level to provide more reliable fault tolerant services for cloud application systems and their tenants. The experimental results show that the proposed method could guarantee the revenue of service providers, and achieve more flexible and more reliable fault tolerant services for cloud application systems with multiple tenants.

Key words: cloud application system, virtual machine deployment, fault tolerance as a service, more reliable

中图分类号: