Research And Implementation Of Key Technologies For Data Center Coverage Area Network Quality Of Service Assessment

The assessment of network service quality in data center coverage areas is an emerging area of research.Studies focusing on the analysis and evaluation of public internet quality between data centers and the clients within their service coverage areas are still in their early stages.In 2019,Microsoft Azure proposed a combined active and passive measurement approach for monitoring data center coverage area networks.However,this approach is tightly coupled with Microsoft’s own infrastructure and may not be applicable to other cloud service providers.Therefore,there is a practical need for research work aimed at developing a universally applicable method for assessing the service quality of data center coverage area networks.This article presents an initial exploratory study on methods for evaluating network service quality in data center coverage areas.The primary research contributions are as follows:(1)A method for evaluating the feasibility of data center coverage area networks is introduced.The study delves into the feasibility aspects of data center coverage areas,examining parameters such as path continuity and path coverage.Simultaneously,it analyzes the strengths and limitations of different levels of granularity in the paths.The research establishes a methodology based on the combination of "AS(Autonomous System)and geographic location information"as the criterion for dividing nodes in the paths,laying the groundwork for the rational partitioning of measurement data sets.(2)Proposal of a Quantile-Based Anomaly Detection Method for Data Center Coverage Area Networks:Data center coverage area networks consist of a vast number of nodes with diverse types and various types of access networks.Relying solely on a single or a few statistical parameters is insufficient for accurately characterizing their service quality characteristics and identifying anomalies.This article introduces a quantile-based method for detecting anomalies in data center coverage area networks,which can extract valuable signals from background noise and enhance the precision of network anomaly detection.(3)Design and Development of Data Center Coverage Area Network Service Quality Evaluation Methods:This research work involves the design,development,and testing of methods for evaluating network service quality in data center coverage areas.The proposed methods are validated based on existing measurement data.