计算机与现代化

• 网络与通信 • 上一篇    下一篇

基于任务特征的云计算资源分配策略

  

  1. (成都理工大学工程技术学院电子信息与计算机工程系,四川乐山614000)
  • 收稿日期:2016-11-24 出版日期:2017-07-20 发布日期:2017-07-20
  • 作者简介:蒋维成(1977-),男,湖南永州人,成都理工大学工程技术学院电子信息与计算机工程系讲师,硕士,研究方向:计算机网络; 李兰英(1979-),女,山东菏泽人,讲师,硕士,研究方向:计算机网络,物联网。
  • 基金资助:
    四川省教育厅科研基金资助项目(16ZB0412); 成都理工大学工程技术学院基金资助项目(C122015007)

Resource Allocation Strategy in Cloud Computing Based on Task Characteristics

  1. (Department of Electronic Information and Computer Engineering, Engineering & 
    Technical College of Chengdu University of Technology, Leshan 614000, China)
  • Received:2016-11-24 Online:2017-07-20 Published:2017-07-20

摘要: 云计算中存在大量的服务资源,高效利用这些资源是资源分配的主要任务。通过对任务流的资源占用和执行过程建模分析,掌握任务流特性,提出伙伴关系任务流资源分配方案。多任务流执行过程中出现资源需求增减的时间同步,这种现象持续不断将产生波动共振,本文采用资源申请-仲裁机制来解决这一问题,把共振的任务流转变为伙伴关系任务流来协作使用资源,保持资源的高效利用。实验结果表明,本文算法消耗的资源比静态方式要少,溢出数也要少,实现了资源的充分利用,验证了算法的正确性。

关键词: 云计算, 任务特征, 资源分配, 共振, 伙伴关系

Abstract: There are a lot of resources in cloud computing. Efficient use of resources is a main challenge of resource allocation in cloud computing. Based on task flow modeling, the resource occupation and task flow execution are analyzed, and the characteristics of the flow are mastered. A resource allocation scheme on partnership task flow is given. In the process of multi-task flow, the demand of resources may increase or decrease at the same time, which will lead to the phenomenon of the wave resonance. The application-arbitration mechanism is proposed to solve this problem, in which the task flow is transformed into the partnership for using resources efficiently. Results show that the resource consumption and overflow of the algorithm are less than that of the static method. The correctness of the algorithm is verified.

Key words:  cloud computing, task characteristics, resource allocation, resonance, partnership

中图分类号: