PerfMon: Measuring Application-Level Performance in a Large-Scale Campus Wireless Network

作     者：Weizhen Dang Tao Yu Haibo Wang Jing’An Xue Fenghua Li Jilong Wang Weizhen Dang;Tao Yu;Haibo Wang;Jing'An Xue;Fenghua Li;Jilong Wang

作者机构：Institute for Network Sciences and CyberspaceTsinghua UniversityBeijing 100084China Department of Computer Science and TechnologyTsinghua UniversityBeijing 100084China Beijing National Research Center for Information Science and Technology(BNRist)Beijing 100084China Huawei TechnologiesBeijing 100085China Zhongguancun LaboratoryBeijing 100094China 

出 版 物：《China Communications》 (中国通信（英文版）)

年 卷 期：2023年第20卷第3期

页      面：316-335页

核心收录：

学科分类：080904[工学-电磁场与微波技术] 0810[工学-信息与通信工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 080402[工学-测试计量技术及仪器] 0804[工学-仪器科学与技术] 081001[工学-通信与信息系统] 

基　　金：supported by the National Key Research and Development Program of China(No.2020YFE0200500) 

主　　题：WiFi traffic patterns network manage-ment performance measurement network diagnosis 

摘      要：WiFi has become one of the most popular ways to access the ***,in large-scale campus wireless networks,it is challenging for network administrators to provide optimized access quality without knowledge on fine-grained traffic characteristics and real network *** this paper,we implement PerfMon,a network performance measurement and diagnosis system,which integrates collected multi-source datasets and analysis *** on PerfMon,we first conduct a comprehensive measurement on application-level traffic patterns and behaviors from multiple dimensions in the wireless network of T university(TWLAN),which is one of the largest campus wireless *** we systematically study the application-level network *** observe that the application-level traffic behaviors and performance vary greatly across different locations and device *** performance is far from satisfactory in some *** diagnose these problems,we distinguish locations and device types,and further locate the most crucial factors that affect the *** results of case studies show that the influential factors can effectively characterize performance changes and explain for performance degradation.