Cost-effective designs of WDM optical interconnects

Optical communication, in particular, wavelength-division-multiplexing (WDM) technique, has become a promising networking choice to meet ever-increasing demands on bandwidth from emerging bandwidth-intensive computing/communication applications, such as data browsing in the World Wide Web, multimedia conferencing, e-commerce, and video-on-demand services. As optics becomes a major networking media in all communications needs, optical interconnects will inevitably play an important role in interconnecting processors in parallel and distributed computing systems. We consider a cost-effective design of WDM optical interconnects for current and future generation parallel and distributed computing and communication systems. We first categorize WDM optical interconnects into two different connection models based on their target applications: the wavelength-based model and the fiber-link-based model. Most of existing WDM optical interconnects belong to the first category. We then present a minimum cost design for WDM optical interconnects under wavelength-based model by using sparse crossbar switches instead of full crossbar switches in combination with wavelength converters. For applications that use the fiber-link-based model, we show that network cost can be significantly reduced, and present such a minimum cost design for WDM optical interconnects under this model. Finally, we generalize the idea used in the design for the fiber-link-based model to WDM optical interconnects under the wavelength-based model, and obtain another new design that can trade off switch cost with wavelength converter cost in this type of WDM optical interconnect. The results in this paper are applicable to any emerging optical switching technologies, such as SOA-based and MEMS-based technologies.     

光互连技术及其在计算机视觉中的应用

可重构的光互连技术具有针对不同的应用提供相应的网络结构的能力。在大规模并行系统中具有很大的优势,可以作为电互连的替代技术应用于计算机视觉这一类计算机和通信速度很高的系统中。     

CMOS Photonics for High-Speed Interconnects

For the first time, high-speed optical communications directly between silicon die are possible at a price-performance point competitive with traditional electrical interconnects.     

Computers and communications: a symbiotic relationship

The communications and computing industries are both enduring a difficult period brought on by too much success in conveying their joint potential to investors and the broader public. The relationship between these two industries, which begin almost at the birth of the modern computing industry, has grown stronger and closer over the years. Just as computing devices are becoming more communications-driven, communication networks are becoming more computing-centric. Computer-enabling services such as broadband and high-speed wireless data will succeed or fail based on how well they connect many millions of people in useful ways. That's something the communications carriers have been doing for 125 years.     

构建基于Windows和MPI的Beowulf并行计算系统

利用普通微机构建并行计算集群(常称为Beowulf系统),能够以低廉的价格获得强大的计算能力。文章介绍了利用16台微机构建一套Beowulf并行计算系统,节点微机上运行Windows2000操作系统,采用MPI(Message-Passing-Interface)的MPICH最新版本:MPICH.NT1.2.3作为并行计算的支撑环境,并以100Mbps高速交换式以太网作为互连网络。通过编制的并行计算程序对该Beowulf系统进行了并行效率的实际测试,测试结果表明该Beowulf系统能够达到非常高的并行加速比和并行效率。     

Free-space interconnects for high-performance optoelectronicswitching

The authors describe the two architectures they have developed that use and demonstrate free-space optical interconnects for digital logic: a high-performance optoelectronic computing module and a second-generation digital optoelectronic computer. The low-power, high-performance optoelectronic computing (HPOC) module was designed for switching and data processing applications. Its architecture uses global, free-space, smart optical interconnects rather than electronic interconnects. The HPOC modules, which incorporate arrays of microlasers, diffractive optical interconnect elements and detectors, are currently undergoing prototype fabrication. By extending the high fan-in and fan-out capabilities of free-space optical interconnects, HPOC modules-when properly configured-offer a significant reduction in power, while maintaining good algorithmic efficiency and high noise margin     

基于多块结构网格大规模并行计算的负载均衡设计及实现

结构网格具有网格生成速度快、质量高、数据结构相对简单、较适用于流体表面应力集中的运算等优点。在大规模 CFD（Computational Fluid Dynamics）并行计算中,需要将网格区域划分为多块网格,而多块网格之间的数据通信会制约并行计算能力的提高,因此对结构网格的负载平衡优化是提高并行计算能力的重点。本文提出了一个采用多层次 k-way 多约束条件图剖分算法来改进负载平衡的方案,并对 M6 翼型和 CRM 模型的多种规模进行了实际计算,结果证明多层次 k-way图剖分算法能够有效地优化负载平衡,在此基础上得到了最优节点间的计算负载平衡和通信负载平衡,最终达到了理想的并行效率。     

A light-weight secure information transmission and device control scheme in integration of CPS and cloud computing

Recently, cloud computing and cyber-physical system (CPS) are definitely basic elements in real industrial field. In particular, security is a mandatory factor for communications and operations in these environments. However, the existing CPS security mechanism is not suitable to the telecommunication framework provided by the standards. In addition, random number function of high entropy must be used to enhance security with encrypted communications and must support perfect secrecy. Random number functions supported by the devices instead of servers do not have sufficient entropy. Entropy injection and seed replacement are also impractical. In this paper, we propose a security scheme which provides light-weight secure CPS information transmission and device control scheme in integration of CPS and cloud computing. In this scheme, a light-weight security scheme can multicast event information to users who have heterogeneous device information access authorities based on oneM2M standards, and also be able to manage the control devices. This paper provides performance analysis of proposed scheme and confirms its security and efficiency.     

Multi-hop communications on wireless network-on-chip using optimized phased-array antennas

Network-on-Chip (NoC) as a promising design approach for on-chip interconnect fabrics could overcome the energy as well as synchronization challenges of the conventional interconnects in the gigascale System-on-Chips (SoC). The advantages of communication performance of traditional wired NoC will no longer be continued by the future technology scaling. Packets that travel between distant nodes of a large scale wired on-chip network significantly suffer from energy dissipation and latency due to the routing overhead at each hop. According to the International Technology Roadmap for Semiconductors annual report, the RFCMOS characteristics will be steadily improved by technology scaling. As the operating frequency of RF devices increases, the size of Si integrated antenna will decrease and it is feasible to employ them as a revolutionary interconnect for intra-chip wireless communications. In this paper, we focus on physical requirements and design challenges of wireless NoC. It is demonstrated that employing an optimum-radiation phased array antenna and multihop communications will increase the reliability of on-chip wireless links by several orders of magnitude using a limited power budget less than 0.1 pJ/bit.     

移动数据库技术和移动计算中的数据管理

移动设备和无线通信技术的快速发展,使得适应移动计算环境的新一代分布式数据库系统,即移动数据库系统迅速成为数据库领域新的研究方向。就移动数据库的基本概念和关键技术加以介绍,并介绍了几种移动计算中数据管理的机制。     

基于Modbus协议的发电机微机励磁实验装置内部通信方法及实现

简单、可靠地实现发电机微机励磁实验装置内部模块之间的数据传递,关系着整套实验装置的稳定性和可靠性。而选用明确、合理的通信协议又是该装置通信安全和快捷的关键所在。作为工业控制领域常用的通信协议——Modbus,有方便易用,性价比高,传输高速、高效等优点。简要介绍了发电机微机励磁实验装置的结构与功能,以及Modbus协议的基本内容。在分析本装置内部通信数据的基础上,分别给出了主机和从机的通信程序。系统运行结果表明,本发电机微机励磁实验装置采用Modbus协议安全、迅速、可靠地实现了主机和从机间的相互通信,既很好地满足了该实验装置的功能要求,又有效地降低了装置的生产成本,具有很好的应用前景。     

The National Applied Mathematics Laboratories-A Prospectus

A prospectus is herein set forth for a federal center of applied mathematics, to be operated as a division of the National Bureau of Standards of the Department of Commerce. The center is to be known as the National Applied Mathematics Laboratories. It will specialize in numerical and statistical analysis, and will undertake to offer various services in these fields, and carry on a broad program of research and training. Particular emphasis will be placed on the development of high-speed automatic computing machinery and the mathematical theory needed for its effective use. (1947)     

A knowledge graph-based data representation approach for IIoT-enabled cognitive manufacturing

The Industrial Internet of Things (IIoT) interconnects a large number of interconnected sensors, actuators, and edge computing devices in the manufacturing systems, where the massive data collected in the manufacturing process has the characteristics of multi-dimensional, heterogeneous, and time series. An effective data representation manner, which can fuse such complex information and enable cognitive manufacturing decision-making from a global perspective, is necessary and challenging. To solve this issue, this paper proposes a knowledge graph-based data representation approach for IIoT-enabled cognitive manufacturing and applies it in a Cyber-Physical Production System (CPPS) scenario. Based on the digital thread of manufacturing process data, a multi-layer manufacturing knowledge graph is established, including device sensing data, production processing data, and business processing data. With the established knowledge graph, a cognition-driven approach is proposed with a perception-cognition dual system, which achieves perception analysis and cognition decision-making in the resource allocation of the manufacturing process. Finally, responding to the orders of personalized products in a workshop is taken as an illustrative example. The performance of allocating resources of workshop devices under dynamic demand changes shows the advantages of the proposed approach. The proposed manner will lay the foundation for a human-like cognition for processing massive real-time industrial information in CPPS, thus paving a pathway towards the era of cognitive manufacturing.     

Virtual network transport protocols for Myrinet

Bringing direct and protected network multiprogramming into mainstream cluster computing requires innovations in three key areas: application programming interfaces, network virtualization systems, and lightweight communication protocols for high-speed interconnects. The AM-II API extends traditional active messages with support for client-server computing and facilitates the construction of parallel clients and distributed servers. Our virtual network segment driver enables a large number of arbitrary sequential and parallel applications to access network interface resources directly in a concurrent but fully protected manner. The NIC-to-NIC communication protocols provide reliable and at-most-once message delivery between communication endpoints. The NIC-to-NIC protocols perform well as the number of endpoints and the number of hosts in the cluster are scaled. The flexibility afforded by the underlying protocols enables a diverse set of timely research efforts. Other Berkeley researchers are actively using this system to investigate implicit techniques for the coscheduling of communicating processes, an essential part of high-performance communications in multiprogrammed clusters of uni- and multiprocessor servers. Other researchers are extending the active message protocols described here for clusters of symmetric multiprocessors, using so-called multiprotocol techniques and multiple network interfaces per machine     

Smart devices and spaces for pervasive computing

Applications and services for pervasive computing have been dramatically grown and have contributed extensively to our daily experiences in recent years. Smart systems, devices, and spaces are proactive for ubiquitous and pervasive computing. Smart information technology (IT) is also an outcome of the state of the art and novel mobile and ubiquitous computing technologies that include highly capable handheld device, pervasive and personal device, etc. This special issue will be a trigger for further related research and technology improvements in pervasive and ubiquitous computing using smart devices and services. This special issue called for original papers describing the latest developments, trends, and solutions of smart devices and spaces for pervasive computing including real-time operating systems (OS), tiny OS and middleware supports, mobile system performance, trustworthy Internet and communications, agents and mobile and pervasive services, among others. In particular, this special issue focuses on a remote control and media-sharing system, flash storage-based smart system, heterogeneous mobile OS, and prediction and auto-execution system for pervasive computing.     

基于USB2．0总线的高速数据采集系统设计

结合当前电荷耦合器件（CCD）信号高速采集面临的问题和USB总线的突出优点,采用USB2．0接口芯片EZ—USBFX2系列CY7C68013A作为USB控制器,复杂可编程逻辑器件（CPLD）EPM7128S为控制核心,外接高速先进先出（FIFO）存储器及16位高速A／D转换模块,设计实现了一个高速数据采集系统。详细介绍了硬件、软件设计。与传统设计相比,该系统具有采集速度快、采样精度高等特点。     

基于硅光子的片上光互连技术研究

随着芯片半导体工艺的发展,芯片集成度不断提高,单个芯片上所能容纳的计算核心数越来越多,使得核心间的数据移动效率成为制约处理器芯片整体性能的关键因素。光互连技术采用波导方式传输数据,信号传输的损耗低、速度快、延迟小,它通过采用波分复用(WDM)技术可以达到很高的带宽密度,有助于解决片上通信的瓶颈问题。面向未来片上高性能互连的需求,深入分析了电互连技术的现状与局限性,研究并分析了基于硅光子的光互连技术发展现状和趋势,对比了多种典型光互连架构的特点及优缺点,总结了未来硅光子互连技术需要解决的5个重要问题。     

Mobile IP与MPLS集成的研究

MPLS技术结合了三层转发的灵活和二层交换的高速,为骨干网络解决方案提供了高速的IP转发能力和极大的可扩展性。MobileIP为用户提供了移动处理的功能。论文介绍了MobileIP与MPLS结合的几种形式,详细研究了在移动节点handoff时,集成方案的处理过程。分析了各种现有的结合方式中的一些缺陷,提出了相应的解决方案。最后简要介绍了MobileIPv6与MPLS集成的方案,并说明这两种技术集成的优势。     

BlueCube: Constructing a hypercube parallel computing and communication environment over Bluetooth radio systems

In parallel computing structures, Hypercubes [P. J. Wan, L. W. Liu, Y. Yang, Optimal routing based on the super-topology in Hypercube WDM networks, 1999, pp. 142–149] and [Y. R. Leu, S. Y. Kuo, A fault-tolerant tree communication scheme for hypercube systems, IEEE Trans. Comput. 45(6) (1996) 643–650] have many advantages: they support parallel computing, provide disjoint paths, and tolerate faults. If devices with computing capabilities can be linked as a Hypercube by taking advantage of Bluetooth radio's features, then an efficient communication and high-performance computing environment can be established by applying currently used algorithms. A Bluetooth device randomly searches for and connects with other devices, using time-consuming inquiry/inquiry scan and page/page scan operation and hence, results in an uncontrolled scatternet topology and inefficient communications. The present work proposes a three-stage distributed construction protocol for rapidly organizing a Hypercube computing environment that was constructed from Bluetooth devices. The proposed protocol governs the construction of links, the assigning of roles and the formation of the scatternet in order to efficiently construct a Hypercube structure. The constructed scatternet easily enables Bluetooth devices to establish a routing path, tolerate faults and create disjoint paths, and thus, achieves parallel and distributed computing in a Bluetooth wireless environment. Experimental results reveal that the proposed protocol can set up a scatternet that is appropriate for parallel computing and communications.     

Microbenchmark performance comparison of high-speed cluster interconnects

Today's distributed and high-performance applications require high computational power and high communication performance. Recently, the computational power of commodity PCs has doubled about every 18 months. At the same time, network interconnects that provide very low latency and very high bandwidth are also emerging. This is a promising trend in building high-performance computing environments by clustering - combining the computational power of commodity PCs with the communication performance of high-speed network interconnects. There are several network interconnects that provide low latency and high bandwidth. Traditionally, researchers have used simple microbenchmarks, such as latency and bandwidth tests, to characterize a network interconnects communication performance. Later, they proposed more sophisticated models such as LogP. However, these tests and models focus on general parallel computing systems and do not address many features present in these emerging commercial interconnects. Another way to evaluate different network interconnects is to use real-world applications. However, real applications usually run on top of a middleware layer such as the message passing interface (MPI). Our results show that to gain more insight into the performance characteristics of these interconnects, it is important to go beyond simple tests such as those for latency and bandwidth. In future, we plan to expand our microbenchmark suite to include more tests and more interconnects.