Container cloud scheduling method based on periodic characteristics

Abstract

The invention relates to a container cloud scheduling method based on periodic characteristics, and belongs to the technical field of cloud computing, and the method comprises the following steps: S1,periodically collecting the system resource usage amount of a container, constructing an online container time domain and frequency domain resource model, and carrying out the analysis to obtain theperiodic characteristics of container resources; s2, according to the periodic characteristics of the container resources, establishing a distributed BFD scheduling strategy to perform container allocation, and obtaining a node with the highest resource utilization rate for each container; and S3, pre-creating a new container at the deployment node, and switching the flow to the new container to complete container migration. Container periodic characteristics are analyzed during container scheduling, container combination scheduling is performed by using the periodic characteristics, and the resource utilization rate of the container cloud platform can be improved while the QoS of the container is ensured; users do not need to intervene in the periodic feature analysis process, and good user friendliness is achieved; and the cluster scheduling responsiveness is improved by using a distributed BFD-based scheduling system.

Description

technical field [0001] The invention belongs to the technical field of cloud computing, and relates to a container cloud scheduling method based on cycle characteristics. Background technique [0002] Container technology includes application packaging, application operating environment packaging, application resource limitation, resource isolation and other technologies. Packaging and packaging the entire runtime through container technology can quickly deploy and run applications in different environments. Compared with traditional virtual machine technology, the host kernel is shared between containers, which improves performance and resource utilization efficiency. Developers can easily copy their production environment as a ready-to-run container application, and operation and maintenance personnel can also use the container provided by the developer for direct deployment, reducing the communication costs caused by different development, testing, and production environm...

AI Technical Summary

Problems solved by technology

[0005] 1. Most customers cannot accurately judge the peak resource usage of containers in terms of resource requirements, and can only apply for more resources as much as possible, resulting in waste of resources

[0006] 2. The container's demand for resources is dynamically changing rather than always in a peak state, which will lead to a large number of idle resources

[0007] 3. The scheduler does not consider the historical value of the container when overselling resources. It only uses priority to deal with resource conflicts, which will lead to a decline in the QoS of low-priority containers, which will intensify as the proportion of resource overselling increases.

Images