A system and method for portable high-availability deployment and management container cluster

Abstract

The invention provides a transplantable high-available container cluster deploying and managing system and method, and belongs to the field of cloud computing. The container cluster deploying and managing system comprises a container cluster, a main scheduling node selection device, a main and standby switching controller, a scheduler, a controller manager and a management northbound interface server. The container cluster deploying and managing method comprises the following steps of R, deploying a container cluster; S, selecting a main scheduling node, and determining on which control node to start and operate the scheduler and the controller manager by the main scheduling node selection device; and T, selecting main and standby control nodes and controlling the switching of the main and standby control nodes by the main and standby switching controller. The system and method provided by the invention is applicable to all container clusters similar to the Kubernetes cluster framework, the high availability problem of the control node is solved, the management and control of the control client on the container cluster is not influenced if the control node goes wrong or the service on the control node goes wrong.

Description

technical field [0001] The invention relates to the field of cloud computing, in particular to a system and method for deploying a container cluster system to realize a highly available cloud computing system. Background technique [0002] As a next-generation resource virtualization technology to improve the current virtual machine, container virtualization technology has gradually become the development focus of cloud computing companies at home and abroad in recent years. With the rapid development of container technology, container cluster technology has become the research frontier in the field of cloud computing technology. Existing container cluster systems such as figure 1 As shown, there is one control node and several slave nodes in a typical existing container cluster system. Users can send management configuration requests to the management and control nodes through the management and control client. The management northbound interface server in the management a...

AI Technical Summary

Problems solved by technology

Users can send management configuration requests to the management and control nodes through the management and control client. The management northbound interface server in the management and control node can accept the management and configuration requests, and deploy, update and delete the user's application container to the slave nodes according to the request. Existing container cluster system of this kind One of the problems is that the high availability of the control node is not well resolved, that is, if there is a problem with the control node or a service on the control node, it will affect the management and control of the container cluster by the control client

Images