首页> 外文期刊>Computer Communications >Kautz graphs as attractive logical topologies in multihop lightwave networks
【24h】

Kautz graphs as attractive logical topologies in multihop lightwave networks

机译:Kautz图是多跳光波网络中有吸引力的逻辑拓扑

获取原文
获取原文并翻译 | 示例
           

摘要

In this paper, we propose Kautz graphs as attractive logical topologies for multihop WDM lightwave networks. Logical topologies proposed in the literature include the ShuffleNet, De Bruijn graph, Manhattan Street Network, and Hypercube. We show that for a given nodal in-degree and out-degree and average number of hops between stations, a network based on Kautz graphs can support a much larger number of stations than competing topologies (ShuffleNet and De Bruijn networks). Our experimental and analytical results show that despite its larger size (more number of stations), a Kautz graph based network shows similar delay and throughput performance as a De Bruijn graph based network. Further, a topology based on Kautz graph has more uniform optical channel loading and utilizes the installed network capacity more efficiently than a De Bruijn network. In addition, fault tolerance properties of Kautz graphs are better than those of De Bruijn graphs'. To deal with non-uniform and asymmetric traffic, an adaptive routing scheme is proposed for Kautz networks, which relieves the network of congestion bottlenecks resulting from non-uniform traffic. Finally, we give the VLSI design and layout a router for Kautz graph-based networks using OASIS (Open Architecture Silicon Implementation Software).
机译:在本文中,我们提出了Kautz图作为多跳WDM光波网络的有吸引力的逻辑拓扑。文献中提出的逻辑拓扑包括ShuffleNet,De Bruijn图,Manhattan Street Network和Hypercube。我们表明,对于给定的节点进站度,出站度以及站点之间的平均跳数,基于Kautz图的网络比竞争拓扑(ShuffleNet和De Bruijn网络)可以支持更多的站点。我们的实验和分析结果表明,尽管基于Kautz图的网络规模较大(站点数量更多),但其延迟和吞吐量性能却与基于De Bruijn图的网络相似。此外,与De Bruijn网络相比,基于Kautz图的拓扑具有更均匀的光信道负载,并且可以更有效地利用已安装的网络容量。另外,Kautz图的容错特性优于De Bruijn图的容错特性。为了处理非均匀和非对称流量,提出了一种针对Kautz网络的自适应路由方案,可以缓解网络中由于非均匀流量导致的拥塞瓶颈。最后,我们使用OASIS(开放式架构芯片实施软件)为VLSI设计和布局用于基于Kautz图的网络的路由器。

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