Design And Implementation Of Cloud Liaison Center Service Monitoring And Management Platform

With the popularity of container technology in cloud computing,more and more enterprises use container technology to encapsulate services into standard container Images for deployment.This approach can greatly improve deployment efficiency,but also make full use of cloud computing resources.Traditional service deployment requires the installation of various environments on which services depend in the operating system,followed by the planning of working contents and monitor ports to prevent conflicts with other services.This has resulted in a lengthy deployment process that is difficult to migrate and expand.The above problems can be solved by using container technology.Containerization is more efficient than other virtualization techniques.The cloud liaison centre is a multimedia multichannel agent system that includes dozens of service components.These components are heavily deployed and prone to errors during the deployment process.In this context,this paper proposes a container technology to realize the easy deployment of cloud communication center system,a service management monitoring platform with good expansibility and stability as well as advanced operations capability.First of all,all services need to be containerized,and the configuration management module is designed to solve the problem that the configuration of the service can not be read after the configuration of the service.To manage these containers,the container choreography technique Kubernetes is introduced and service deployment module is developed based on Kubernettes.At the same time,in order to monitor the operation situation of host machine and service instances in the cluster,a service monitoring module is designed and implemented,which monitor the hardware resource footprint such as CPU,memory on host and the resources occupied by a single container process,and statistics the completion indicators of the service business function.Although the load balancing function is provided in Kubernetes,the policy is relatively single and can not manage the traffic in the cluster more finely.To solve this problem,this paper focuses on the design of traffic management module based on the expansion of kubernetes core components.This paper innovatively proposes a traffic agent model that combines performance and security;adopts a design architecture that is completely decoupled with the operation,achieves operation non-perception;Provide user identity-based traffic guidance function;and carries out a special handling for the websocket protocol.In the test section,the final test results show that the deployment management function is running normally,and the function indexes of the traffic management module also meet the requirements.At present,the platform has been deployed in the school's enrollment consultation system and has been officially launched before the 2018 college entrance examination registration.At the end of the paper,the research work and shortcomings of this paper are summarized,and the future research work is prospected.