A container resource planning system and method based on application image data identification

Abstract

The invention discloses a container resource planning system and method based on application image data identification. The system includes a container management platform, a container resource planning module, and one or more applications, and each application includes two forms of an application image and an application container. , the application image and the application container are provided with one or more; application information probes are deployed on the container management platform; resource model library, resource model calculation engine and resource planning controller are deployed on the container resource planning module; The method includes steps 1-7. The present invention uses the overall container image data as the basis for identifying the resource consumption type of the application container, comprehensively analyzes the resource consumption type of the application container through rich application information probes, and applies the machine learning technology to the application. In the analysis and identification of container resource consumption types, the accuracy of application type identification and resource configuration is greatly improved.

Description

technical field [0001] The invention relates to resource planning for applications on a container platform, in particular to a container resource planning system and method based on application image data identification. Background technique [0002] Containers are widely used due to their characteristics of second-level startup, lightweight, and security isolation. In the process of using container platforms, it is necessary to manage resources for applications, including the CPU, memory, network, process, and storage used by application containers. / O, etc., usually set a limit value and a reserved value for each resource, the limit value is the upper limit of the resources that the container can use, and the reserved value is the lower limit of the resources used by the container, that is, the system reserves The resource share, by setting the limit value and reserved value, realizes performance isolation and efficient use of resources, prevents resource waste and ensures...

AI Technical Summary

Problems solved by technology

[0011] The technical problem to be solved by the present invention is that the number of containers on the container platform may be huge, especially in the micro-service scenario, so the workload of manual setting is very large and repetitive, and it is extremely error-prone , the present invention provides a container resource planning system based on application image data identification, using the overall container image data as the basis for identification of application container resource consumption types, and comprehensively analyzing application container resource consumption types through rich application information probes, And machine learning technology is applied to the analysis and identification of application container resource consumption types, which greatly improves the accuracy of application type identification and the accuracy of resource allocation; it can automatically identify the container type to be deployed on the container platform according to the container image data And classification, and based on the policy to give the recommended value of the corresponding container resources, which greatly improves the efficiency and accuracy; the whole process is fully automated, and a large number of application resource consumption models based on best practices are preset, which greatly reduces the It reduces the workload and threshold of manual container resource configuration, improves the accuracy of container resource configuration, and is very suitable for the current more and more extensive containerized application deployment scenarios; it can also collect and analyze real-time running data of applications. The application container resource model is continuously optimized to further improve the scientificity of subsequent application resource planning through continuous self-learning to solve the defects caused by existing technologies

Images