电信科学 ›› 2023, Vol. 39 ›› Issue (2): 71-82.doi: 10.11959/j.issn.1000-0801.2023028

• 研究与开发 • 上一篇    下一篇

基于Kubernetes的多云网络成本优化模型

高明, 刘铭, 陈泱婷, 王伟明   

  1. 浙江工商大学信息与电子工程学院,浙江 杭州 310018
  • 修回日期:2023-02-09 出版日期:2023-02-20 发布日期:2023-02-01
  • 作者简介:高明(1979- ),男,博士,浙江工商大学信息与电子工程学院副教授、网络系主任,主要研究方向为新型网络体系架构和工业互联网
    刘铭(1997- ),男,浙江工商大学信息与电子工程学院硕士生,主要研究方向为新型网络体系架构和云原生网络
    陈泱婷(1998- ),女,浙江工商大学信息与电子工程学院硕士生,主要研究方向为软件定义网络
    王伟明(1964- ),男,博士,浙江工商大学信息与电子工程学院教授,主要研究方向为新一代网络体系结构和开放可编程网络
  • 基金资助:
    国家自然科学基金资助项目(61871468);浙江省基础公益研究计划项目(LGG20F010015);浙江省新型网络标准与应用技术重点实验室基金资助项目(2013E10012)

Cost optimization model for multi-cloud network based on Kubernetes

Ming GAO, Ming LIU, Yangting CHEN, Weiming WANG   

  1. School of Information and Electronic Engineering, Zhejiang Gongshang University, Hangzhou 310018, China
  • Revised:2023-02-09 Online:2023-02-20 Published:2023-02-01
  • Supported by:
    The National Natural Science Foundation of China(61871468);The Basic Public Welfare Research Program of Zhejiang Province(LGG20F010015);The Key Laboratory of Network Standards and Applied Technology Foundation of Zhejiang Province(2013E10012)

摘要:

以 Kubernetes 为代表的云原生编排系统在多云环境中被云租户广泛使用,随之而来的网络观测性问题愈发突出,跨云跨地区的网络流量成本尤为突出。在Kubernetes中引入扩展的伯克利数据包过滤器(extended Berkeley packet filter,eBPF)技术采集操作系统内核态的网络数据特征解决网络观测问题,随后将网络数据特征建模为二次分配问题(quadratic assignment problem,QAP),使用启发式搜索与随机搜索组合的方法在实时计算的场景下求得最佳近优解。此模型在网络资源成本优化中优于 Kubernetes 原生调度器中仅基于计算资源的调度策略,在可控范围内增加了调度链路的复杂度,有效降低了多云多地区部署环境中的网络资源成本。

关键词: Kubernetes, eBPF, 多云网络, 二次分配问题

Abstract:

The cloud-native scheduling system, represented by Kubernetes, is widely used by cloud tenants in a multi-cloud environment.The problem of network observation becomes more and more serious, especially the cost of network traffic across cloud and region.In Kubernetes, the eBPF technology was introduced to collect the network data features of kernel state of operating system to solve the network observation problem, and then the network data features were modeled as QAP, a combination of heuristic and stochastic optimization was used to obtain the best near optimal solution in a real-time computing scenario.This model is superior to the Kubernetes native scheduler in the cost optimization of network resources, which is based on the scheduling strategy of computing resources only, and increases the complexity of scheduling links in a controllable range, effectively reduces the cost of network resources in a multi-cloud area deployment environment.

Key words: Kubernetes, eBPF, multi-cloud network, quadratic assignment problem

中图分类号: 

No Suggested Reading articles found!