通信学报 ›› 2012, Vol. 33 ›› Issue (Z1): 192-200.doi: 10.3969/j.issn.1000-436x.2012.z1.025

• 学术论文 • 上一篇    下一篇

云应用引擎的资源监控和计费机制研究

任怡1,张菁1,陈红1,吴庆波1,孔金珠1,戴华东1,管刚2   

  1. 1 国防科学技术大学 计算机学院,湖南 长沙 410073
    2 腾讯研究院,北京 100080
  • 出版日期:2012-09-25 发布日期:2017-08-03
  • 基金资助:
    国家高技术研究发展计划(“863”计划)基金资助项目;国家科技支撑计划基金资助项目;国家自然科学基金资助项目

Research on resource monitoring and billing mechanisms of application engine in cloud computing environment

Yi REN1,Jing ZHANG1,Hong CHEN1,Qing-bo WU1,Jin-zhu KONG1,Hua-dong DAI1,Gang GUAN2   

  1. 1 College of Computer Science,National University of Defense Technology,Changsha 410073,China
    2 Tencent Research Institute,Beijing 100080,China
  • Online:2012-09-25 Published:2017-08-03
  • Supported by:
    The National High Technology Research and Development Program of China (863 Program);The National Science and Technology Support Program Foundation of China;The National Natural Science Foundation of China

摘要:

在归纳分析已有云计费机制的基础上,选取开源的云应用引擎 AppScale 作为研究对象,分析了其组成与架构,探讨了其资源监控功能的实现机理和不足。修改并扩展了 AppScale 的资源监控代码,设计实现了云应用引擎的资源监控和计费机制CloudMB,该机制支持进程级资源监控和面向多租户的计费功能。经测试,CloudMB占用系统CPU时间不超过3%,对平台性能影响较小。

关键词: 云应用引擎, 平台即服务, 资源监控, 多租户, 计费

Abstract:

A comprehensive study on the billing methods of existing cloud computing platforms was made.Then,AppScale,an open sourced scalable and typical PaaS software,as the foundation platform of our target resource monitoring and billing mechanisms was chose.Appscale’s structure,its components and how they worked together was analyzed.CloudMB,which enhanced AppScale by supporting process level resource monitoring and multi-user oriented billing function with flexible billing policies was designed and implemented.Finally,experiment verified that CloudMB is effective and its occupation of CPU time is lower than 3%.

Key words: application engine, PaaS, resource monitor, multi-tenant, billing

No Suggested Reading articles found!