针对异构融合网络中不同网络的Qo S和映射问题,解决分类和映射方法是提出Qo S参数映射模型。在充分研究现有异构网Qo S参数映射方法的基础上,提出了Qo S参数映射三维模型(Three-dimensional Model for Qo S Parameter Mapping,Qo S-TMP...针对异构融合网络中不同网络的Qo S和映射问题,解决分类和映射方法是提出Qo S参数映射模型。在充分研究现有异构网Qo S参数映射方法的基础上,提出了Qo S参数映射三维模型(Three-dimensional Model for Qo S Parameter Mapping,Qo S-TMPM),模型采用颗粒度适中、区分度较大的Qo S分类关键参数进行搭建。同时,基于Qo S参数映射三维模型的Qo S参数映射方法能顺利地完成异构网络间的Qo S参数映射,仿真结果表明Qo S参数模型和参数映射方法能改善Qo S保证服务,从而提高用户的业务体验质量。展开更多
As one of Qo S(Quality of Service) metrics, delay is critical important to delay sensitive applications, such as interactive video, network game and online surgery. In this paper, we exploit SDN(Software Defined Netwo...As one of Qo S(Quality of Service) metrics, delay is critical important to delay sensitive applications, such as interactive video, network game and online surgery. In this paper, we exploit SDN(Software Defined Networking) advantages to solve delay Qo S problem. Our work mainly focuses on SDN based queuing delay estimation with real traffic and end-to-end delay control. First, we propose a queuing estimation model and extended it for end-to-end delay of the whole path. It is proved to be feasible and accurate with experimental results in SDN environment. Second, in order to demonstrate the use of our proposed model, we also implement an end-to-end delay control application in SDN. It fulfills specific delay Qo S requirements by dynamically switching flows to a suitable queue based on estimation results and delay requirements.展开更多
基金supported by the project of 111 Intelligence Introduction for Innovation at Communication University of ChinaNSFC under grant No.61371191,No.61472389 and No.61201236ETH Zurich for the valuable discussion about H2020 MAMI project related to TCP modeling in this work
文摘As one of Qo S(Quality of Service) metrics, delay is critical important to delay sensitive applications, such as interactive video, network game and online surgery. In this paper, we exploit SDN(Software Defined Networking) advantages to solve delay Qo S problem. Our work mainly focuses on SDN based queuing delay estimation with real traffic and end-to-end delay control. First, we propose a queuing estimation model and extended it for end-to-end delay of the whole path. It is proved to be feasible and accurate with experimental results in SDN environment. Second, in order to demonstrate the use of our proposed model, we also implement an end-to-end delay control application in SDN. It fulfills specific delay Qo S requirements by dynamically switching flows to a suitable queue based on estimation results and delay requirements.
