A container group lifecycle management method based on Kubernetes system architecture

Abstract

The invention discloses a life cycle management method of a container group based on a Kubernetes system architecture, comprising the following steps: 1, a user submits a task to a Kubernetes system for application processing; 2, the user submits the task to the Kubernetes system for application processing. 2, that Kubernetes system judges whether there is a pod correspond to the current task or not, if there is no pod corresponding to the current task, the pod is created, and a timed destruction task is set for the pod; 3. Before the scheduled destruction task of the Pod is executed, whetherthe task staging pool for tasks that correspond to the current Pod is checked, If there is a task corresponding to the current Pod, the task corresponding to the current Pod in the task temporary storage pool is imported into the Pod, the original timing destruction task of the Pod is canceled, and a timing destruction task is set for the temporary Pod again; and if there is a task corresponding to the current Pod, the task corresponding to the current Pod in the task temporary storage pool is imported into the Pod. In order to avoid the problem of Pod being destroyed and rebuilt constantly inthe scenario of frequent task submission, the mode of destruction countdown after Pod execution is adopted, which effectively improves the actual utilization rate of Pod and reduces the risk of out-of-control system resources.

Description

technical field [0001] The invention belongs to the technical field of Kubernetes system architecture application, and in particular relates to a container group lifecycle management method based on the Kubernetes system architecture. Background technique [0002] Kubernetes is a brand-new leading solution for distributed architecture based on container technology. It is a popular application method to use Kubernetes to manage container groups and perform specific tasks that require an isolated environment in container groups. Container groups provide unified services externally in the form of Pod objects, and Pods are the basic unit for hosting applications or services. [0003] In the actual application process, developers need to pay attention to the life cycle of the Pod, which means that the developer must specify when the application or service hosted in the Pod needs to be created and destroyed. A reasonable Pod lifecycle management strategy will help to effectively ...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is that when the Kubernetes system processes temporary tasks that require isolation from each other, these temporary tasks do not need to maintain a service for a long time, and the use of the existing task-insensitive life cycle management strategy is likely to cause container group Pod The problem of wasting system resources in the idle state for a long time provides a container group lifecycle management method based on the Kubernetes system architecture

Images