A Kubernetes-Based Monitoring Platform for Dynamic Cloud Resource Provisioning
A Kubernetes-Based Monitoring Platform for Dynamic Cloud Resource Provisioning
复制标题
基于 Kubernetes 的动态云资源配置监控平台
DOI：
10.1109/glocom.2017.8254046
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
Jeu
中科院分区：
文献类型：
--
作者：
Chia;Shun;En;P. Lin;Jeu
最近，越来越多的网络运营商已经部署云环境来实现网络运营中心，该网络运营中心监控其大规模移动的或有线网络的状态。通常，云环境采用基于容器的虚拟化，使用Docker进行容器打包，使用Kubernetes进行多主机Docker容器管理。在这样一个基于容器的环境中，Kubernetes可以动态监控资源需求和/或正在运行的应用程序的使用情况，然后相应地调整提供给托管容器的资源，这一点很重要。目前，Kubernetes提供了一种简单的动态资源配置机制，该机制只考虑CPU利用率，因此并不有效。本文旨在开发一个通用平台，以促进基于Kubernetes的动态资源配置。我们的平台包含以下三个功能。首先，我们的平台包括一个全面的监测机制，集成和提供相对完整的系统资源利用率和应用程序的QoS指标的资源配置算法，使更好的供应策略。其次，我们的平台模块化的动态资源配置操作的操作，使用户可以很容易地部署一个新设计的算法，以取代我们的平台中现有的。第三，在我们的平台中的动态资源配置操作被实现为一个控制循环，因此可以应用到所有运行的应用程序后，用户定义的时间间隔，而无需其他手动配置。
Recently, more and more network operators have deployed cloud environment to implement network operations centers that monitor the status of their large-scale mobile or wireline networks. Typically, the cloud environment adopts container-based virtualization that uses Docker for container packaging with Kubernetes for multihost Docker container management. In such a container-based environment, it is important that the Kubernetes can dynamically monitor the resource requirements and/or usage of the running applications, and then adjust the resource provisioned to the managed containers accordingly. Currently, Kubernetes provides a naive dynamic resource-provisioning mechanism which only considers CPU utilization and thus is not effective. This paper aims at developing a generic platform to facilitate dynamic resource-provisioning based on Kubernetes. Our platform contains the following three features. First, our platform includes a comprehensive monitoring mechanism that integrates and provides the relatively complete system resource utilization and application QoS metrics to the resource-provisioning algorithm to make the better provisioning strategy. Second, our platform modularizes the operation of dynamic resource- provisioning operation so that the users can easily deploy a newly designed algorithm to replace an existing one in our platform. Third, the dynamic resource-provisioning operation in our platform is implemented as a control loop which can consequently be applied to all the running application following a user-defined time interval without other manual configuration.