Casimer M. DeCusatis

Wireless optical transmission of fast ethernet, FDDI, ATM, and ESCON protocol data using the TerraLink laser communication system

The TerraLink laser communication (lasercom) system was developed as a cost-effective, high-bandwidth, wireless alternative to fiber optic transmission. The advantages of lasercom over fiber optic cabling are primarily economic. However, free-space lasercom is subject to atmospheric effects, such as attenuation and scintillation, which can reduce link availability and may introduce burst errors not seen in fiber transmission. The TerraLink transceivers use large receive apertures and multiple transmit beams to reduce the effects of scintillation. By designing the lasercom link with sufficient margin for atmospheric attenuation and scintillation, a bit error rate (BER) of 1029 or better can be achieved. Since we designed the TerraLink transceivers to be eye-safe at the transmit aperture, each system is range-limited. Link power budgets for the TerraLink systems are presented, and link margin data are shown that quantitatively describe how the effective laser link range varies in different weather conditions. Since the TerraLink transceivers act as simple repeaters, they are protocol-independent. Examples of TerraLink installations transmitting wireless fast ethernet (125 Mbits/s), fiber distributed data interface (FDDI) (125 Mbits/s), asynchronous transfer mode (ATM) (155 and 622 Mbits/s), and Enterprise Systems Connection (ESCON) (200 Mbits/s) protocol data are presented.

Communication within clouds: open standards and proprietary protocols for data center networking

Cloud computing and other highly virtualized data center applications have placed many new and unique requirements on the data center network infrastructure. Conventional network protocols and architectures such as Spanning Tree Protocol and multichassis link aggregation can limit the scale, latency, throughput, and virtual machine mobility for large cloud networks. This has led to a multitude of new networking protocols and architectures. We present a tutorial on some of the key requirements for cloud computing networks and the various approaches that have been proposed to implement them. These include industry standards (e.g., TRILL, SPB, software-defined networking, and OpenFlow), best practices for standards-based data center networking (e.g., the open datacenter interoperable network), as well as vendor proprietary approaches (e.g., FabricPath, VCS, and Qfabric).

Software defined networking to support the software defined environment

Software defined networking (SDN) represents a new approach in which the decision-making process of the network is moved from distributed network devices to a logically centralized controller, implemented as software running on commodity servers. This enables more automation and optimization of the network and, when combined with software defined compute and software defined storage, forms one of the three pillars of IBMs software defined environment (SDE). This paper provides an overview of SDN, focusing on several technologies gaining attention and the benefits they provide for cloud-computing providers and end-users. These technologies include (i) logically centralized SDN controllers to manage virtual and physical networks, (ii) new abstractions for virtual networks and network virtualization, and (iii) new routing algorithms that eliminate limitations of traditional Ethernet routing and allow newer network topologies. Additionally, we present IBMs vision for SDN, describing how these technologies work together to virtualize the underlying physical network infrastructure and automate resource provisioning. The vision includes automated provisioning of multi-tier applications, application performance monitoring, and the enabling of dynamic adaptation of network resources to application workloads. Finally, we explore the implications of SDN on network topologies, quality of service, and middleboxes (e.g., network appliances).

Parallel optical interconnects for enterprise class server clusters: needs and technology solutions

We examine the status of enterprise-class server clusters and the communication issues that need to be addressed in future systems. With increasing system performance, new approaches beyond traditional copper-only communication solutions have to be examined. Parallel optics is an attractive solution to overcome coppers shortcomings, but traditional approaches to parallel optics have had their own limitations. We describe a new approach to parallel optics - dense parallel optics - and its relevance to enterprise servers. After discussing system communication needs, we examine dense parallel optics from both the passive cabling and active component views. We explain how dense parallel optics offers a unique way to address the performance, cost, reliability, and scalability of server systems. We also discuss how current approaches to dense parallel optics afford system opportunities beyond simple data transport.

Optical Interconnect Networks for Data Communications

In this invited tutorial paper, we review the changing nature of data center networks and the role played by optoelectronics in future network designs. Conventional network protocols will be reviewed, including Ethernet, Fibre Channel, and InfiniBand, and requirements for WAN connectivity between data centers. The transition to converged networks based on lossless Ethernet will be discussed, including FCoE and RoCE protocols. Industry roadmaps for bandwidth, port density, and scalability of optical links will be presented, and optical transceiver form factors including QSFP, CFP, and active optical cables will be discussed. The role of software defined networking (SDN) in next generation data center networks will also be presented in this context.

Performance comparison of small form factor fiber optic connectors

A comparison of Small Form Factor (SFF) fiber optic connectors is presented for LC, MT-RJ, SC-DC, and VF-45. Multimode and singlemode jumper cables were tested using industry standard test procedures and bench marked against the industry standard SC Duplex connector. Initial loss data as well as stress testing was performed. Variations were found in the performance of the connectors types and between connectors of the same type from different suppliers. In some cases the connectors out performed the SC Duplex on some tests but no connectors out performed the SC Duplex on all tests with the mechanical stress tests of axial and off-axial pull being the most difficult to exceed. These connectors are rapidly developing and are at different levels of maturity but none of them tested out to be a fully mature replacement for the SC Duplex connector yet.

Spread-spectrum techniques in optical communication using transform domain processing

A method of applying spread-spectrum techniques to optical communication is presented. The interference-suppression capability of spread-spectrum systems is shown to be enhanced by optical transform domain processing, and a design for spectral coding at optical frequencies is given. The encoding system is based on optical pulse compression and shaping. Several possible implementations of this system are suggested, and applications to fiber optics, laser radar, free-space optical communications, and other systems are discussed. >

IBM system z10 I/O subsystem

The performance, reliability, and functionality of a large server are greatly influenced by the design characteristics of its I/O subsystem. The critical components of the IBM System z10™ I/O subsystem have, therefore, been significantly improved in terms of performance, capability, and cost. The first-order network has been redesigned from the long-evolved enhanced self-timed interface (eSTI) links to utilize InfiniBand™ links. A redesign of the host logic of I/O chips and the fiberoptic interfaces within the links made it possible to introduce InfiniBand-based IBM Parallel Sysplex® links. A broad range of legacy I/O channels have been carried forward to connect through InfiniBand, and a foundation has been laid for new channel types of improved functionality and performance. The first such hardware channel to be introduced is the next generation of Ethernet-virtualization data routers. A new and methodical recovery structure has been designed to ensure consistent, extensive support of reliability, availability, and serviceability. A building-block-oriented design process has been developed to enable the innovations that made these advances possible. Finally, a new performance verification methodology has been introduced to ensure that the system and subsystem designs are balanced to make effective use of the increased capacity.

Fiber optic cable infrastructure and dispersion compensation for storage area networks

The fiber optic cable infrastructure currently used by many large data centers is over 10 years old, and may soon require upgrading in order to support higher-data-rate services. Legacy multimode optical fiber is typically low bandwidth, and will only support limited distances at data rates exceeding 1 Gb/s. In this article we investigate various methods for extending the distance of multimode fiber at higher data rates. This includes the tactical use of enhanced bandwidth multimode fiber, electronic dispersion compensation, and wavelength tuning control loops.

R3C2: Reactive Route and Rate Control for CEE

In order to fully realize the potential of Cloud and High Performance Computing (HPC) applications, significant improvement is required in the cost/performance of data center networks. While recent industry standards such as Quantized Congestion Notification (QCN) for Converged Enhanced Ethernet (CEE) have begun to address this issue, there are still significant problems left open. Therefore we propose two novel source-based adaptive routing schemes for CEE-based networks. First, we develop a basic source-driven Reactive Route Control (R2C2) adaptive routing scheme. In response to congestion notifications, the source activates additional paths to re-route traffic around potential congestion points. Using industry standard VLANs, a source node can effectively control the path choices in the network. This approach goes beyond conventional QCN limitations by replacing its reaction point with a VLAN-based multipath route controller. We thus enable HPC/Cloud applications demanding direct and/or secure access to the network features. Second, we combine R2C2 with the QCN reaction point, resulting in the higher performance Reactive Route & Rate Controller (R3C2). In case of persistent or multiple hotspots when VLAN route selection alone is insufficient, the R3C2 source will throttle its packet injection rates individually along each congested route of a multipath bundle. Detailed simulations against established data center and HPC benchmarks show the practical benefits in performance and stability.