Method for improving container cloud resource expansion efficiency

Abstract

The invention particularly relates to a method for improving container cloud resource expansion efficiency. The method for improving the container cloud resource expansion efficiency comprises using declarative management nodes and component resources to deploy components in a distributed and concurrent mode; hierarchically scheduling nodes an performing resource scheduling by taking an application unit as a unit; adding a Kubernetes component for traffic control, and hierarchically accessing data; pulling a mirror image layer file by utilizing a P2P technology, and loading a container as required when using; reducing the network delay by using a BGP mode; and realizing container network rule matching in an IPVS mode. According to the method for improving the container cloud resource expansion efficiency, the cluster deployment time can be shortened, the Node expansion efficiency can be improved, the container scheduling delay, the Kubernetes resource processing delay and the network delay can be reduced, so that the storage filing capability, the cluster high availability, the container starting speed and the Pod expansion efficiency are improved, the problem that large-scale cluster Service scale matching is slow is solved, and the benchmark performance, the long-term stability performance and the expansion performance of the large-scale hydrogen cluster platform are guaranteed.

Description

technical field [0001] The present invention relates to the technical field of container expansion network, and in particular to a method for improving the expansion efficiency of container cloud resources. Background technique [0002] Cloud computing technology has experienced the development process from physical machine to virtual machine and from virtual machine to container. Container technology is already a relatively mature technology for providing microservices. With the expansion of service scale, the scale of container clusters is getting larger and larger. A container cloud cluster manages hundreds or even thousands of machines. As the size of the cluster increases, the efficiency of container orchestration and scheduling becomes lower and lower. How to improve the expansion efficiency of container cloud resources is a challenge faced by various cloud vendors. [0003] Kubernetes (K8S for short, container orchestration and scheduling system) is a leading distribu...

AI Technical Summary

Problems solved by technology

When the performance of the ApiServer component is improved, the Controller (container controller component) component cannot perceive the changes of the ApiServer in a timely manner; all components of Kubernetes eventually fall to the Etcd (container metadata access database) storage, and Etcd has high read and write delays. Insufficient expansion efficiency of Node, insufficient expansion efficiency of Pod, and excessive network performance loss

[0006] To sum up, the problems of large-scale clusters are specifically manifested as: cluster operation commands are stuck, nodes are unavailable, and container creation takes a long time

Images