首页> 外文学位 >Hybrid adaptation layers for high-performance optical packet switched IP networks.
【24h】

Hybrid adaptation layers for high-performance optical packet switched IP networks.

机译:用于高性能光分组交换IP网络的混合适配层。

获取原文
获取原文并翻译 | 示例

摘要

With the increasing demand for network bandwidth, today's electronic routers struggle to keep up and are fast reaching their physical limitations. All-Optical packet switching offers an alternative approach to forwarding traffic by keeping the payload in the optical domain throughout the network while forwarding is done based solely on the optical header. Advantages of such an approach include a scalable increase in payload bit rate, packet forwarding rate and reduction in power consumption and physical size.; Current research in optics has yielded several integrated and non-integrated technologies leading towards all-optical packet forwarding. Research on an adaptation mechanism that achieves complete payload and frame decoupling and thereby enabling direct IP over optics packet forwarding has however been limited. Additionally, core packet forwarding involves statistical multiplexing of packet traffic. Contention is a major issue during statistical multiplexing and has been conventionally solved using Random Access Memory (RAM) buffers in electronics. In the absence of Optical RAMs, a new technique of handling contention using fixed length delay lines as buffering is required. This thesis focuses on research of a new optical framing mechanism that enables all-optical statistical demultiplexing of IP over optics at the core and optical processing mechanisms such as packet compression and decompression to enable all-optical packet forwarding by statistical multiplexing at the network core. Research on efficient adaptation of slow speed IP traffic at the edge to enable and improve forwarding at the high speed optical core is presented in this work.
机译:随着对网络带宽需求的增加,当今的电子路由器努力跟上并迅速达到其物理极限。全光分组交换通过将有效载荷保持在整个网络的光域中,而仅基于光头进行转发,提供了一种转发流量的替代方法。这种方法的优点包括有效载荷比特率,分组转发率的可扩展增加以及功耗和物理尺寸的减小。当前的光学研究已经产生了几种集成和非集成技术,这些技术导致了全光分组转发。然而,对实现完全有效负载和帧解耦并由此实现直接IP over光学分组转发的自适应机制的研究受到了限制。此外,核心数据包转发涉及数据包流量的统计复用。争用是统计复用期间的主要问题,并且通常已使用电子产品中的随机存取存储器(RAM)缓冲区解决。在没有光学RAM的情况下,需要一种使用固定长度的延迟线作为缓冲来处理竞争的新技术。本文的重点是研究一种新的光成帧机制,该机制能够在核心上实现对IP over IP的全光统计多路分解,以及诸如分组压缩和解压缩之类的光学处理机制,从而能够通过在网络核心进行统计多路复用来进行全光分组转发。在这项工作中,提出了在边缘有效地适应慢速IP流量以实现和改善高速光核心的转发的研究。

著录项

  • 作者

    Rangarajan, Suresh.;

  • 作者单位

    University of California, Santa Barbara.;

  • 授予单位 University of California, Santa Barbara.;
  • 学科 Engineering Electronics and Electrical.
  • 学位 Ph.D.
  • 年度 2006
  • 页码 195 p.
  • 总页数 195
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 无线电电子学、电信技术;
  • 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