Container scheduling method associated with time based on Kubernetes container cluster

Abstract

The invention discloses a container scheduling method associated with time based on a Kubernetes container cluster, and belongs to the technical field of cloud computing. According to the method, whether judgment is carried out in a time range or not is added between a pre-selection strategy and an optimal selection strategy of a scheduler of the Kubernetes cluster; calculating the average load score of the pre-selected available Node. According to the technical scheme provided by the invention, whether the scheduling operation is further carried out or not can be decided by comparing the scheduling operation with the preset score, so that the normal operation of the existing business application is ensured, and some tasks with time correlation can better utilize resources in the cluster.For example, in the morning at night, According to the technical scheme of the invention, when the load of the business application is large, the time period is the time period with the load of most business applications being not heavy, the time period is very suitable for some big data offline computing tasks, the tasks are allocated to the Node where the business application is located within the time period, the operation efficiency of the big data tasks can be improved, and the utilization rate of hardware resources in the cluster can be improved.

Description

technical field [0001] The invention relates to a software resource scheduling method, in particular to a time-related container scheduling method based on a Kubernetes container cluster, and belongs to the technical field of cloud computing. Background technique [0002] Kubernetes, referred to as K8s, is an abbreviation formed by replacing the eight characters "ubernete" with 8. It is a distributed architecture platform based on container technology. Kubernetes provides container applications with functions such as service registration, load balancing, service deployment and operation, service rolling upgrade, online expansion and contraction, resource scheduling, resource quota management, etc. It can be said that Kubernetes has complete cluster management capabilities and runs through distributed systems. Development, testing, deployment, operation and maintenance monitoring all links. [0003] The method of scheduling and managing software resources based on Kubernetes...

AI Technical Summary

Problems solved by technology

[0005] 2) Resource management HPA based on software performance indicators, such as adjusting the number of pods according to the CPU and memory usage of the currently running pods, realizes automatic expansion and contraction functions, but it cannot accurately arrange pods on specific Nodes, Therefore, pod resources cannot be effectively scheduled according to the characteristics and time of specific services

It means that the latter's intensive application does not have a Node running the pod of the former's computing-intensive application, not only cannot effectively schedule pod resources, but also causes resources such as CPU, memory, and I / O to not be properly allocated to Needed applications, resulting in waste of some hardware resources

Images