Container cluster delayed shrinkage scheduling method and system

Abstract

The invention discloses a container cluster delayed shrinkage scheduling method and system and belongs to the container cluster shrinkage scheduling field. The method comprises the following steps: 1) an agent monitoring module monitors in real time performance index and health status data of a container bearing a service and sends the result to a scheduling module; 2) the scheduling module judges whether the reported performance index and the health status data surpass preset threshold values, if so, sends an instruction to the agent monitoring module to create a new container, or otherwise, judges whether there is a newly-created container, and if there is the newly-created container, turn to the step 3); and 3) the scheduling module judges whether the performance index is smaller than the preset threshold value, if so, a load balancer stops allocating service traffic, and after the data is processed, the scheduling module issues an instruction to delete the new container, or otherwise, turn to the step 1). The whole system is prevented from the detect of data loss of an existing mode through delay and TCP resetting, thereby ensuring that the cases of packet loss and brief interruption of services are not caused in the load balancing configuration update process to the maximum degree.

Description

technical field [0001] The present invention relates to the field of container cluster shrinkage scheduling, in particular to a container cluster delay shrinkage scheduling method and system. Background technique [0002] At present, there are many open source application container engines, such as docker, which allow developers to package their applications and dependencies into a portable container, and then publish them to any popular Linux machine, and can also realize virtualization, docker It is a "process" in the cluster world; docker has become an ideal container technology choice for applications that need to run across multiple different environments. Through docker, we can easily manage services and dependencies between services. [0003] After the generation of docker, most enterprises began to use containers to build services in order to better save resources. However, when the services encounter some peak periods, if there are too few containers or the CPU memo...

AI Technical Summary

Problems solved by technology

[0004] Based on the above technical problems, the present invention provides a delay shrinkage scheduling method and system for container clusters, thereby solving the technical problems of internal data loss and temporary interruption of services caused by directly stopping containers in existing dynamic shrinkage. The delay shrinkage scheduling method of the container cluster and the system use delay to effectively avoid the shortcomings of data loss and temporary service interruption caused by directly stopping the container

Images