GOALI-FIND: Optical Flow Switched Core Networks

近年来，光流交换作为一种很有前途的新方法出现在光网络中，以满足指数级增长的流量需求。为了实现光网络的这种新范式，核心网络必须能够在接近一秒或更快的时间尺度上进行波长重构和业务提供。目前用于波分复用（WDM）光传输的方法是针对静态波长通道设计的，这些通道的数量可以随时间增加，并且可以在配置时配置形成不同的路径。提供新的波长是一个有条不紊、循序渐进的过程。出现?光信号沿路径传输，需要几分钟到几个小时才能达到最终的稳定状态，适合GMPLS网络。此外，关于光传输的大量文献假设了这种准静态网络环境。该项目的智力价值将在于确定物理层约束对光流交换网络施加的实际和基本限制，包括动态波长交换环境中的光传输。由于问题的多层性质，这个目标特别具有挑战性。网络拓扑和信令协议将规定物理层传输要求，例如光信号必须以多快的速度达到无差错状态，以及传输距离。同样，性能和成本方面的物理限制将激励架构和协议的设计选择。为了解决光流交换网络的多层特性，该GOALI项目被提议作为一种机制，利用贝尔实验室的新型光学透明网状网络测试平台设施，扩展麻省理工学院NSF资助的NeTS-FIND未来光网络架构项目中的网络架构活动。特别地，我们将结合MIT的光流交换网络优化研究，在测试台上进行有针对性的实验。这个新的测试平台最近被用于研究可重构光网络中的功率动力学和信号传输性能。这两个元素对于理解透明交换网络中的物理层是必不可少的，该设施的独特之处在于它将这些元素结合在一个宽带、长途传输配置中，这是典型的核心网络。这种校企合作将在许多层面产生更广泛的影响。扩展光网络以服务急剧增长的互联网流量的问题具有广泛的社会影响。该项目将使麻省理工学院和贝尔实验室两家协同研究机构之间的合作成为可能，以解决这一重要问题。学术参与者将有机会在一流的工业研究实验室工作，并受益于该环境特有的知识和技能。在此过程中，该计划将提供一个独特的机会，将两个组织内促进个人驱动的传统融合在一起

Optical flow switching has recently emerged as a promising new approach to address the exponentially growing traffic demands in optical networks. In order to realize this new paradigm for optical networks, the core network must be capable of wavelength reconfiguration and service provisioning on time scales approaching one second or faster. The present methods used in wavelength division multiplexed (WDM) optical transmission are designed for static wavelength channels, which can increase in number over time and may be configurable to form different paths at the time of provisioning. Provisioning new wavelengths is a methodical, one at a time, step-wise process of ?turning up? the optical signal along the path, which requires minutes to hours in order to achieve the final stable state, appropriate for GMPLS networks. Furthermore, the extensive literature on optical transmission assumes this quasi-static network environment. The intellectual merit of this project will be to determine both the practical and fundamental limitations on optical flow switched networks imposed by physical layer constraints, including optical transmission in a dynamic, wavelength switching environment. This goal is particularly challenging because of the multi-layer nature of the problem. The network topology and signaling protocols will dictate physical layer transmission requirements, such as how fast the optical signal must achieve an error free condition and over what transmission distance. Likewise the physical limitations both in terms of performance and cost will motivate design choices for the architectures and protocols. In order to address the multi-layer nature of optical flow switched networks, this GOALI project is proposed as a mechanism to utilize the novel optically transparent mesh network testbed facility within Bell Laboratories to extend the network architecture activities within the NSF funded NeTS-FIND Future Optical Network Architectures program at MIT. In particular, targeted experiments will be conducted in the testbed in conjunction with the optical flow switched network optimization studies at MIT. This new testbed has recently been used to study both the power dynamics and the transmission performance of signals in reconfigurable optical networks. These two elements are essential for understanding the physical layer in a transparent switched network and this facility is unique in that it combines these in a broadband, long haul transmission configuration, typical of core networks. This university-industry collaboration will have broader impact at many levels. The problem of scaling optical networks to serve the dramatic rise in internet traffic has broad social ramifications. The project will enable cooperation between two synergistic research efforts to address this important problem: at MIT and at Bell Labs. The academic participants will have the opportunity to work in a premier industrial research laboratory and benefit from the knowledge and skills specific to that environment. In the process, the program will provide a unique opportunity to merge the traditions within both organizations of promoting individual driven