首页> 外文期刊>Applied optics >Optimized design of delay-line buffers with an input-feedback mechanism for asynchronous optical packet switching networks
【24h】

Optimized design of delay-line buffers with an input-feedback mechanism for asynchronous optical packet switching networks

机译:具有异步光分组交换网络的输入反馈机制的延迟线缓冲器的优化设计

获取原文
获取原文并翻译 | 示例
           

摘要

The delay-line buffer is widely used to solve contention by exploiting the time domain for optical packet switching (OPS) networks. How to optimize the length of the delay-line element to minimize the packet loss is a major challenge in designing delay-line buffers efficiently. In this paper we consider the buffers in asynchronous OPS networks with different incoming traffic patterns. For a specific mean packet length, we first demonstrate that the optimum length value of the delay-line element mainly depends on the incoming traffic load. The dependence of packet loss ratio on the delay-line granularity under different traffic loads and buffer sizes is given by both analytical model and simulation results. An optimum delay granularity is found under each traffic load, and this optimum granularity is independent of the buffer sizes. According to this finding, we designed an adaptive delay-line buffer structure with an input-feedback mechanism, namely an improved multiple-input single-output optical buffer, which could automatically configure the switch fabric and adjust the delay-line granularity according to the incoming traffic load. It is shown that the proposed buffer structure can achieve optimum performance for packet loss rate under different network scenarios. (C) 2016 Optical Society of America
机译:延迟线缓冲器通过为光分组交换(OPS)网络开发时域而广泛用于解决竞争。如何有效地设计延迟线缓冲器的长度是最大的挑战,这是有效设计延迟线缓冲器的主要挑战。在本文中,我们考虑异步OPS网络中具有不同传入流量模式的缓冲区。对于特定的平均数据包长度,我们首先证明延迟线元素的最佳长度值主要取决于传入的业务负载。分析模型和仿真结果均给出了不同流量负载和缓冲区大小下丢包率对延迟线粒度的依赖性。在每个流量负载下找到最佳延迟粒度,并且该最佳粒度与缓冲区大小无关。根据这一发现,我们设计了一种具有输入反馈机制的自适应延迟线缓冲器结构,即一种改进的多输入单输出光缓冲器,该缓冲器可以自动配置交换结构并根据延迟调整粒度。传入流量负载。结果表明,所提出的缓冲器结构可以在不同的网络情况下实现最佳的丢包率性能。 (C)2016美国眼镜学会

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