Wavelength-division Multiplexing (WDM) Optical Interconnect Architectures for Parallel and Distributed Computing and Communications

用于并行和分布式计算和通信的波分复用 (WDM) 光互连架构

• 批准号: 0207999
• 负责人: Yuanyuan Yang
• 金额: --
• 依托单位: SUNY at Stony Brook
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2002
• 资助国家: 美国
• 起止时间: 2002-09-01 至 2007-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0207999&HistoricalAwards=false
• 关键词: Wavelength; division; Multiplexing; WDM; Optical

Optical communication, in particular, wavelength-division multiplexing (WDM) technique, has become a promising networking choice to meet ever-increasing demands on bandwidth from many emerging bandwidth-intensive computing/communication applications. As optics become a major networking media, optical interconnects will inevitably play an important role in interconnecting processors in parallel/distributed computing systems. This research focuses on fundamental challenges and issues on using optics in two converging areas: parallel/distributed computing and communications. The objective of this research is to design high-speed, cost-effective optical interconnects for current and future generation parallel/distributed computing and communication systems. Due to the unique characteristics of optics, many important issues of optical interconnects, different from those of electronic interconnects, need to be addressed. Specially, this research focuses on (1) explore unique properties and classifications of WDM interconnects; (2) optimal and cost-effective designs; (3) study the effect of wavelength conversion; (4) performance modeling; (5) fault-tolerance issues. The proposed research combines architecture/circuit design, algorithmic, probabilistic, combinatorial and simulation techniques to conduct comprehensive studies on the above issues. The research results will provide viable solutions for designing high-bandwidth, high-connectivity, low-latency and low-cost optical interconnects. The proposed research will have a significant impact on next generation arallel/distributed computing and communication systems and future networking infrastructure.

光通信，特别是波分复用（WDM）技术，已经成为满足许多新兴的带宽密集型计算/通信应用对带宽的不断增长的需求的有前途的联网选择。随着光学成为主要的网络介质，光学互连将不可避免地在并行/分布式计算系统中的处理器互连中发挥重要作用。 这项研究的重点是在两个融合领域使用光学的基本挑战和问题：并行/分布式计算和通信。本研究的目的是为当前和未来的并行/分布式计算和通信系统设计高速，低成本的光互连。 由于光学器件的特殊性，光学互连有许多不同于电子互连的重要问题需要解决。本研究的重点是：（1）探索WDM互连的独特属性和分类;（2）最佳和成本有效的设计;（3）研究波长转换的效果;（4）性能建模;（5）容错问题。 建议的研究结合架构/电路设计，算法，概率，组合和仿真技术，对上述问题进行全面的研究。 研究结果将为设计高带宽、高连接性、低延迟和低成本的光互连提供可行的解决方案。拟议的研究将对下一代集群/分布式计算和通信系统以及未来的网络基础设施产生重大影响。