Method and device for realizing elastic expansion and contraction of number of cross-cluster containers

Abstract

The invention provides a method and a device for realizing elastic expansion and contraction of number of cross-cluster containers. The method comprises the following steps: calling an API (Application Program Interface) of a scaling service instance in a leader role to configure an application needing scaling and a corresponding scaling strategy; the instance executing the scaling task obtains the monitoring index of the application in the cluster from the monitoring service, weighted averaging is carried out according to the number of copies, and a comprehensive index value is obtained through calculation; executing an instance of the scaling task, and when the comprehensive index value is not in a preset scaling threshold interval, calculating a reasonable copy number in each cluster according to a preset weight of each cluster and a maximum value and a minimum value of the copy number; and executing the instance of the scaling task, calling the API of the scaling strategy, and scaling the copy number of the to-be-processed application in the corresponding cluster. According to the invention, under the condition that the disaster recovery requirement is met, the availability of the service system is improved, and the application system is ensured to always have proper capacity so as to meet the current traffic demand.

Description

technical field [0001] This application relates to the field of cloud-native technology, and in particular, to a method and device for implementing elastic scaling of cross-cluster containers based on a distributed architecture in a cloud-native Kubernetes environment. Background technique [0002] With the advancement of enterprise digital transformation, cloud native has become an important engine driving business growth. As the de facto industry standard for cloud-native applications, Kubernetes (hereinafter referred to as k8s) provides an efficient unified management mechanism for containerized deployment, overall planning, construction and upgrade of enterprise resource applications. [0003] At present, the simultaneous deployment and operation of k8s multi-clusters is inevitable. On the one hand, the hybrid cloud or multi-cloud architecture determines the need for multiple clusters; on the other hand, if there is only one cluster and the cluster fails, all services wi...

AI Technical Summary

Problems solved by technology

[0003] At present, the simultaneous deployment and operation of k8s multi-clusters is inevitable. On the one hand, the hybrid cloud or multi-cloud architecture determines the need for multiple clusters; on the other hand, if there is only one cluster and the cluster fails, all services will be affected.

However, resource scheduling is currently limited by clusters, application availability is limited by clusters, and elastic scaling is limited by clusters. There is no mechanism for elastically scaling the number of containers across clusters.

Images