Method and system for pod scheduling in kubernetes cluster

A cluster and scheduling operation technology, which is applied in the direction of multi-programming devices, program control design, instruments, etc., can solve the problems of high node scheduling pressure on the management plane
CN113268337BActive Publication Date: 2021-10-22深圳软通动力信息技术有限公司
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
深圳软通动力信息技术有限公司
Publication Date
2021-10-22

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The present application relates to a method and system for pod scheduling in a Kubernetes cluster, wherein a plurality of working nodes in the cluster form a unidirectional ring topology, and the plurality of working nodes include a leader node, the leader node and the management plane node Communication, the Pod scheduling method in the Kubernetes cluster includes: the leader node pulls the metadata set from the management plane node, thereby determining the list of Pods to be run, and for each Pod in the list of Pods to be run, the topology of the large node The working nodes in the structure perform Pod scheduling operations in turn until the working node running the Pod is determined. Through this application, the problem of high node scheduling pressure on the management plane in the related technology is solved when the number of working nodes in the Kubernetes cluster is large. problem, achieving the effect of reducing the pressure of node scheduling on the management plane.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present application relates to the technical field of computer software technology application, in particular to a method and system for pod scheduling in a Kubernetes cluster. Background technique

[0002] In a container orchestration engine (Kubernetes, k8s for short) cluster, the scheduler component of the management plane node is responsible for selecting a suitable working node for the Pod to run. The specification of the management plane node needs to be able to support scheduling and management of large-scale working nodes. In terms of technology, the Kubernetes cluster supports a maximum of 5000 working nodes. Within the upper limit of the number of working nodes, in the case of a large number of working nodes in the Kubernetes cluster, the management plane node scheduling pressure is greater. As the number of working nodes increases , the scheduling pressure on the nodes on the management plane is increasing, which may cause the Kubernetes...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More