首页> 外文会议>IEEE AFRICON Conference >Model analysis of dynamic priority segmented burstification and deflection routing
【24h】

Model analysis of dynamic priority segmented burstification and deflection routing

机译:动态优先级分段突发和偏转路由的模型分析

获取原文

摘要

Heterogeneous data traffic is increasingly becoming a complex mix of both delay tolerant and intolerant IP traffic. Optical Burst Switched (OBS) networks have since become a backbone network solution to handle the resultant zetabyte volumes of such traffic concurrently using the burst assembly and transmission approach. The strict minimum burst size constraint requirements as well as varying maximum delay bounds for most traffic types makes it difficult to render an acceptable and consistent quality of service (QoS) for the aggregated traffic in such networks Hence in this paper we propose a dual priority with jump priority segmented burst assembling coupled with a most efficient route (MER) selection when deflection contending bursts in the network interior. At the egress nodes, newly arriving low priority (LP) data segments may under certain conditions, jump to the high priority (HP) queue. In that way the delays are minimized and this leading to improvement in overall renderable QoS. We provide a queuing model analysis of this model as well as the QoS metrics. In the core network, taking into cognizance that deflection routing based contention control can exasperate/accelerate congestion, we propose that the most efficient route (MER) be chosen for deflecting the contending burst, as this will reduce congestion on loaded routes. Analytical as well as simulation results show that dual priority burst segmentation with jump priority approach, coupled with MER deflection routing can improve on QoS metrics while at the same time utilizing the available network resources rationally and efficiently under moderate to high network traffic conditions.
机译:异构数据流量日益成为延迟容忍和不宽容的IP流量的复杂组合。光突发交换(OBS)网络已经从此成为骨干网络解决方案来处理这种业务的所得zetabyte卷同时使用组件中的脉冲串和传输的方法。严格的最小突发尺寸限制的要求,以及不同的最大延迟边界的大部分流量类型,难以使服务于这样的网络聚合业务可接受的和一致的服务质量(QoS),因此,在本文中,我们提出了双优先跳跃优先分段突发装配加上一个最有效的路由(MER)的选择时偏转竞争脉冲串在网络内部。在出口节点,新到达的低优先级(LP)的数据段可在某些条件下,跳转到高优先级(HP)队列。在这样的延迟最小化,这导致整体渲染服务质量的改善。我们提供这种模式的排队模型分析以及对QoS指标。在核心网络中,考虑到认定该偏转路由基于竞争的控制可以恶化/加速拥塞,我们建议,最有效的路由(MER)被选择用于偏转竞争脉冲串,因为这将减少对加载的路由拥塞。分析以及仿真结果表明,双优先爆裂跳跃优先方法,加上MER偏转路由,而在同一时间利用可用的网络资源合理有效地在中等至高网络流量状况可以改善QoS度量分割。

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