{"title":"Network Optimization for Differentiated QoS Traffic in an SDN Environment for PoP-Data Center Traffic","authors":"M. Maswood, Rohit Abhishek, D. Medhi","doi":"10.1109/SARNOF.2018.8720505","DOIUrl":null,"url":null,"abstract":"A major challenge for network providers is to design a network that can cater to differentiated quality-of-service (QoS) for different traffic classes in a cost-effective manner. Traditional differentiated services (“DiffServ”) is for IP networks where differentiation is done in IP packet headers through code points. However, this does not by itself land in architecting a network from a traffic engineering point of view. In an SDN environment, with more control over flows or a collection of flows, differentiated QoS traffic classes can be considered and accordingly a network can be optimized. In this work, we consider network optimization in a software-defined network (SDN) environment where differentiated QoS traffic is provided through different latency bounds. We present a mixed integer linear programming (MILP) formulation for this problem. The study shows that our approach can maintain more stringent QoS requirement by reducing the value of maximum allowable latency without a significant increment in bandwidth cost.","PeriodicalId":430928,"journal":{"name":"2018 IEEE 39th Sarnoff Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 IEEE 39th Sarnoff Symposium","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SARNOF.2018.8720505","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
Abstract
A major challenge for network providers is to design a network that can cater to differentiated quality-of-service (QoS) for different traffic classes in a cost-effective manner. Traditional differentiated services (“DiffServ”) is for IP networks where differentiation is done in IP packet headers through code points. However, this does not by itself land in architecting a network from a traffic engineering point of view. In an SDN environment, with more control over flows or a collection of flows, differentiated QoS traffic classes can be considered and accordingly a network can be optimized. In this work, we consider network optimization in a software-defined network (SDN) environment where differentiated QoS traffic is provided through different latency bounds. We present a mixed integer linear programming (MILP) formulation for this problem. The study shows that our approach can maintain more stringent QoS requirement by reducing the value of maximum allowable latency without a significant increment in bandwidth cost.