Research And Implementation Of Data Service Quality Assurance Method Based On Container Scheduling

With the rapid development of technologies such as cloud computing and the Internet of Things,the types of data resources are becoming increasingly diverse,and the demands of users to process data sources are increasing.Data service refers to an information technology-driven service that provides data collection,data storage,data processing,data exchange,and evolution of various survival forms of data.Due to users' needs for services,the resources that services rely on,and the relationship between services are dynamic and uncertain,ensuring the quality of service has become a basic requirement of the container cloud.Therefore,this article focuses on data service quality assurance issues.There are many factors that affect the quality of service.So first,the article designs a service quality container cloud dynamic monitoring architecture by considering not only the basic resources of carrying services but also the delay caused by the relationship between services or between services and data sources.Therefore,a dynamic monitoring architecture of container cloud for data service quality and a visualization system based on monitoring data are designed.Second,a runtime service scheduling method is proposed in our article.The proposed method first transforms the quality-based service scheduling problem into a planning problem that constrained by basic resource usage and inter-service delay,then effectively generates a service optimization scheduling plan through particle swarm optimization to achieve the ultimate goal of ensuring the quality of data services.Finally,the monitoring architecture and scheduling method are verified and analyzed by experiments.The main research work of this paper is as follows:1.In order to solve the delay sensitivity,volatility,and dependency challenges faced by service quality,a container cloud monitoring architecture for service quality is proposed.The architecture takes into account the monitoring of basic resource usage and service delay.It is composed of a container cluster and a monitoring server.The container cluster is the basic environment of the container cloud and is the object of monitoring.The monitoring server provides various types of monitoring data collection,external access to monitoring data and monitoring response function based on monitoring data.2.In order to improve the single-source service scheduling method,a runtime data-oriented service scheduling method is proposed.This method clarifies the scheduling object,scheduling process,scheduling strategy,and scheduling algorithm.When the monitored index of the service reaches the threshold set in advance,the service is scaled and the new service copies generated by scaling are scheduled.The essence of the scheduling algorithm is to model the scheduling-based service dynamic deployment problem as a constraint-planning problem and use particle swarm algorithm to perform the optimization process.In terms of constraints,it is no longer only based on the utilization of basic resources,but also takes into account the overall resource configuration of the container cloud and the dependencies between services or services and data source to achieve the ultimate goal of ensuring service quality.3.A container cloud monitoring visualization system for data service quality has been designed and implemented,and the feasibility of its monitoring response function has been verified.Combined with the analysis of application cases in the transportation field,the feasibility and effectiveness of the method in this paper are verified through comparative experiments with the original scheduling method of k8 s.The experimental results show that the method can ensure the resource utilization and balance of the container cloud environment at the same time.The delay between services or service and data source is reduced,and the quality of service is effectively guaranteed.