• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE Conference on Local Computer Networks >Deterministic Confidence Interval Estimation of Networking Traffic in SDN
【24h】

Deterministic Confidence Interval Estimation of Networking Traffic in SDN

机译：SDN中网络流量的确定性置信区间估计

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Software Defined Networking (SDN) enables a centralised entity - the controller - to monitor the network's status by collecting traffic statistics such as packets, bytes, etc. Each statistic is associated with a forwarding table entry (FTE) in a switch whose structure and format is specified by the OpenFlow standard (de-facto SDN standard). For a flow with a FTE, its statistic is easily acquired by an inquiry from a controller to the switch on this flow's corresponding FTE. If a flow has no matching FTE, its statistic is not known until a new FTE is installed for the purpose of monitoring it. However, the time to install these FTEs and the potential conflicts between them and the existing FTEs jeopardise the feasibility of this approach. To avoid these drawbacks, this paper proposes a traffic estimation approach based on the existing FTE's statistics. With the help of boolean algebra, the deterministic confidence interval of any given flowset can be estimated. This approach avoids the FTE installation time and also saves the FTE storage space.
机译：软件定义网络（SDN）使集中式实体（控制器）可以通过收集流量统计信息（例如数据包，字节等）来监视网络状态。每个统计信息都与交换机的转发表条目（FTE）相关联，该转发表条目的结构和格式由OpenFlow标准（事实上的SDN标准）指定。对于具有FTE的流，可以通过从控制器查询该流的相应FTE的开关来轻松获取其统计信息。如果流没有匹配的FTE，则在安装新的FTE进行监视之前，未知其统计信息。但是，安装这些FTE的时间以及它们与现有FTE之间的潜在冲突危及了此方法的可行性。为避免这些弊端，本文提出了一种基于现有FTE统计信息的流量估算方法。借助布尔代数，可以估算任何给定流集的确定性置信区间。这种方法避免了FTE的安装时间，还节省了FTE的存储空间。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号