首页> 外文学位 >An improved distributed routing protocol for multi-domain optical networks using topology aggregation.
【24h】

An improved distributed routing protocol for multi-domain optical networks using topology aggregation.

机译:使用拓扑聚合的多域光网络的一种改进的分布式路由协议。

获取原文
获取原文并翻译 | 示例

摘要

Multi-domain networks consist of several domains connected together by inter-domain physical links. Routing and failure protection in such networks are complex issues that have only started to receive attention. One source of difficulties lies in the lack of shared information among domains, such as network topologies, bandwidth, and size of the network. This kind of information can only be estimated across the different domains. Truong and Jaumard in "Using Topology Aggregation for Efficient Shared Segment Protection Solutions in Multi-Domain Networks", IEEE journal, propose a two-step routing and protection approach where in the first step a topology aggregation and cost estimation is obtained to model the inter-domain network followed by a detailed intra-domain routing step. Different approximation techniques are used to estimate costs and residual bandwidths in the aggregated model of the multi-domain network. But protection and routing interact in a complex manner, which renders difficulty in these estimations. In this thesis, we propose a technique that aims at improving the cost estimation in the aggregated topology.;The virtual estimated cost can be improved by computing “real virtual edges” whenever there is an inter-domain request in multi-domain network. The estimated cost is significantly improved by the proposed method in this study though there is some overhead involved. The complete protocol is simulated using Boost Graph Library in C++ environment. The simulation was performed on a large graph and results were compared with original GROS heuristic. From the tests, we can confirm that the heuristic proposed in this thesis will help to achieve better cost estimation.
机译:多域网络由通过域间物理链接连接在一起的几个域组成。此类网络中的路由和故障保护是复杂的问题,仅引起人们的关注。困难的根源之一是域之间缺乏共享信息,例如网络拓扑,带宽和网络大小。这种信息只能跨不同的域进行估计。 Truong和Jaumard在IEEE期刊“将拓扑聚合用于多域网络中的高效共享网段保护解决方案”中,提出了一种两步路由和保护方法,其中第一步是获得拓扑聚合和成本估算以对内部-域网络,然后进行详细的域内路由步骤。在多域网络的聚合模型中,使用了不同的近似技术来估计成本和剩余带宽。但是保护和路由以复杂的方式相互作用,这给这些估计带来了困难。在本文中,我们提出了一种旨在改善聚合拓扑中的成本估算的技术。当多域网络中存在域间请求时,可以通过计算“真实虚拟边缘”来提高虚拟估算成本。尽管涉及一些开销,但是通过本研究中的建议方法可以显着改善估计成本。使用C ++环境中的Boost Graph Library模拟完整的协议。在大图上执行了仿真,并将结果与​​原始GROS启发式算法进行了比较。通过测试,我们可以证实本文提出的启发式方法将有助于实现更好的成本估算。

著录项

  • 作者单位

    Oklahoma State University.;

  • 授予单位 Oklahoma State University.;
  • 学科 Computer Science.
  • 学位 M.S.
  • 年度 2012
  • 页码 49 p.
  • 总页数 49
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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