Search

History of fiber optics / Jeff D. Montgomery -- Market analysis and business planning / Yann Y. Morvan and Ronald C. Lasky -- Small form factor fiber optic connectors / John Fox and Casimer DeCusatis -- Specialty fiber optic cables / Casimer DeCusatis and John Fox -- Optical wavelength division multiplexing for data communication networks / Casimer DeCusatis -- Optical backplanes, board and chip interconnects / Rainer Michalzik -- Parallel computer architectures using fiber optics / David B. Sher and Casimer DeCusatis -- Packaging assembly techniques / Ronald C. Lasky, Adam Singer, and Prashant Chouta -- InfiniBand, the interconnect from backplane to fiber / Ali Ghiasi -- New devices for optoelectronics : smart pixels / Barry L. Shoop, Andre H. Sayles, and Daniel M. Litynski -- Emerging technology for fiber optic data communication / Chung-Sheng Li -- Manufacturing challenges / Eric Maass.

The Handbook includes chapters on all the major industry standards, quick reference tables, helpful appendices, plus a new glossary and list of acronyms. This practical handbook can stand alone or as a companion volume to DeCusatis: Fiber Optic Data Communication: Technological Advances and Trends (February 2002, ISBN: 0-12-207892-6), which was developed in tandem with this book.* Includes emerging technologies such as Infiniband, 10 Gigabit Ethernet, and MPLS Optical Switching* Describes leading edge commercial products, including LEAF and MetroCore fibers, dense wavelength multiplexing, and Small Form Factor transceiver packages* Covers all major industry standards, often written by the same people who designed the standards themselves* Includes an expanded listing of references on the World Wide Web, plus hard-to-find references for international, homologation, and type approval requirements* Convenient tables of key optical datacom parameters and glossary with hundreds of definitions and acronyms* Industry buzzwords explained, including SAN, NAS, and MAN networking* Datacom market analysis and future projections from industry leading forecasters

A one-stop desk reference for R&D engineers involved in communications engineering, this book will not gather dust on the shelf. It brings together the essential professional reference content from leading international contributors in the field. Material covers a wide scope of topics, including voice, computer, facsimile, video, and multimedia data technologies. - A hard-working desk reference, providing all the essential material needed by communications engineers on a day-to-day basis - Fundamentals, key techniques, engineering best practice and rules-of-thumb together in one quick-reference sourcebook - Definitive content by the leading authors in the field

All modern data centers require some form of data backup or replication to protect the data from natural or man-made disasters and provide business continuity. Companies rely on their information systems to run daily operations. If a system becomes unavailable, company operations may be impaired or stopped completely. If critical data remains inaccessible for an extended period, the company may never recover and be forced to go out of business. It is necessary to provide a reliable infrastructure for IT operations in order to minimize any chance of disruption. In this chapter, we define the requirements for Tier 1 through Tier 4 data centers. We discuss the ACID-BASE (atomicity, consistency, isolation, durability-basically available, soft state, eventual consistency) taxonomies for data consistency, giving examples from companies such as Yahoo!, Amazon, Google, and IBM. The chapter includes a detailed discussion of the different options for IBM Geographically Dispersed Parallel Sysplex (GDPS), enterprise-class high-end business continuity, and disaster recovery solution, including the Sysplex Timer protocol, InterSystem Channel (ISC), Parallel Sysplex InfiniBand (PSIFB), and more.

While the modern data center still contains a mixture of different application-specific networks and protocols, various types of Ethernet have emerged as the preferred approach for many data networking problems. With the addition of new industry standard extensions that permit lossless transmission, converged enhanced Ethernet (also known as lossless Ethernet) is now widely available. New applications have emerged, such as the convergence of networking and storage through Fibre Channel over Ethernet, remote direct memory access over Ethernet (iWarp), and the use of Ethernet to supplement or replace conventional interconnects between data centers (metro Ethernet and related efforts). This chapter will provide an overview of Ethernet protocols, including both classic versions and more recent updates targeted for data center applications.

This chapter discusses the networking requirements for a new class of integrated data center appiances that have emerged within the past few years. Conventional data centers employ a combination of servers, storage, networking, software applications, and services, which are usually combined for the first time at the customer site, and which frequently involve building custom solutions for specific jobs. Building custom systems is not a sustainable or economical approach; data center applications can take months to design and deploy, and there are ongoing issues with interoperability, support, scaling, and performance tuning to meet application requirements. Further, the provisioning and configuration of custom solutions is repetitive, manually intensive, and prone to errors. We discuss a new class of factory preintegrated systems combining servers, storage, and networking. Examples of integrated systems are discussed including offerings from Cisco, Hewlett Packard, Hitachi, Huawei, IBM, Oracle/Sun, and others.

In recent years, there have been many fundamental changes in the architecture of modern data centers. New applications have emerged, including cloud computing, big data analytics, real-time stock trading, and more. Workloads have evolved from a predominantly static environment into one that changes over time in response to user demands, often as part of a highly virtualized, multitenant data center. In response to these new requirements, data center networks have also undergone significant change. Conventional network architectures, which use Ethernet access, aggregation, and core tiers with a separate storage area network, are not well suited to modern data center traffic patterns. This chapter reviews the evolution from conventional network architectures into designs better suited to dynamic, distributed workloads. This includes flattening the network, converging Ethernet with storage and other protocols, and virtualizing and scaling the network. Effects of oversubscription, latency, higher data rates, availability, reliability, energy efficiency, and network security will be discussed.

In recent years, there have been many new approaches to data networking protocols, both industry standard and vendor proprietary. In this chapter, we will begin with an overview of conventional networking protocols, such as the Spanning Tree Protocol and multichassis link aggregation, and network design approaches like equal cost multipath spine–leaf. We will then review several more recent proposals for addressing the requirements of a flattened, Layer 2 network infrastructure. We discuss Transparent Interconnection of Lots of Links (TRILL) and Shortest Path Bridging, as well as both industry standard and proprietary network options including Open Data Center Interoperable Network (ODIN), Qfabric, FabricPath, and Virtual Cluster Switching (VCS). Overlays including Virtual Extensible LAN (VXLAN), Network Virtualizaation Generic Routing Encapsulation (NVGRE), Distributed Overlay Virtual Ethernet (DOVE), and others will also be discussed.

This chapter will examine the technical requirements for designing fiber optic data communication systems. We begin with an introduction to basic link budget design techniques, suitable for practitioners such as network designers, followed by a more detailed review of optical link design considerations. Optical link design concepts discussed include transmitter output power, receiver sensitivity, and properties of optical connectors, couplers, and splices. Figures of merit including bit error rate and signal-to-noise ratio are defined. Both worst case and statistical techniques are presented, including test and validation methods for the completed link. Optical power penalties are discussed, including dispersion (modal and chromatic), intersymbol interference, mode partition noise, mode hopping, extinction ratio, relative intensity noise, multipath interference, wavelength-dependent attenuation, bending loss, timing jitter, radiation-induced darkening, modal noise, and stimulated Raman and Brillouin scattering.