Entry Hoffnagle:1995:Pd from ibmsysj.bib

Last update: Thu Nov 27 02:08:39 MST 2008                Valid HTML 3.2!

Index sections

Top | Symbols | Math | A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

BibTeX entry

@Article{Hoffnagle:1995:Pd,
  author =       "Gene F. Hoffnagle",
  title =        "Preface",
  journal =      j-IBM-SYS-J,
  volume =       "34",
  number =       "4",
  pages =        "552--553",
  year =         "1995",
  CODEN =        "IBMSA7",
  ISSN =         "0018-8670",
  bibdate =      "Tue Mar 19 17:38:46 1996",
  note =         "Issue Order No. G321-0122.",
  URL =          "http://www.research.ibm.com/journal/sj34-4.html#one",
  abstract =     "An essential ingredient in the movement of large
                 amounts of data on worldwide digital networks is
                 agreement on and computer support for internationally
                 recognized architectures for broadband communications
                 services. IBM has contributed to the development of
                 international standards for asynchronous transfer mode
                 (ATM), which underlies much of the effort to produce
                 systems for broadband computer communications and
                 digital networks. IBM's high-speed, multimedia
                 networking architecture and products, based on the ATM
                 architecture, are known by the term Networking
                 BroadBand Services (NBBS) and fit under the umbrella of
                 IBM's Open Blueprint*.
                 \par
                 This issue contains an introductory essay on ATM and
                 NBBS and ten papers on the NBBS approach, architecture,
                 technologies, and products. In combination with the
                 previous issue of the Journal on network technologies
                 and systems, this issue shows the direction IBM has
                 taken in delivering architecture, systems, and products
                 for worldwide digital communications. We are indebted
                 to M. R. Wiggins of the IBM Networking Hardware
                 Division in Research Triangle Park, North Carolina, for
                 his contributions to the creation, development, and
                 preparation of this issue.
                 \par
                 The development of high-performance networks is driven
                 by demand for high levels of network integration,
                 transparent connectivity, high capacity, high speed,
                 and lower costs. Cohen and Abensour, in an introductory
                 essay, discuss those needs, describe the ATM
                 architecture as a response to those needs, and present
                 IBM's strategy for fulfilling the ATM architectural
                 requirements.
                 \par
                 Budhiraja et al. show in their paper, multicasting
                 requires an architecture and protocols that support
                 definition and control of the set of recipients and
                 provide high-speed and high-quality network
                 connections. NBBS contains a multicast architecture
                 that meets these requirements, for which the authors
                 supply an overview.
                 \par
                 Ahmadi et al. discuss a further aspect of networks in
                 general and ATM networks in particular: traffic
                 management. Included in this broad term are such
                 important and fundamental tasks as path selection,
                 bandwidth allocation, administration of service
                 guarantees, and various other connection-, packet-, and
                 cell-level controls. In this paper the authors describe
                 the associated NBBS features and their
                 interaction.
                 \par
                 Selection of network paths and links for high-speed
                 and, notably, multimedia applications with
                 quality-of-service guarantees and constraints is the
                 subject of a paper by Tedijanto et al. Since the
                 resulting selection is a nonpolynomial- (NP-) complete
                 problem, the authors describe their heuristic algorithm
                 for determining an effective path. The algorithm has
                 features such as connection and preemption priorities
                 and the use of primary and secondary paths.
                 \par
                 Network control protocols have traditionally assumed
                 relatively slow and error-prone networks. Today's
                 high-speed and reliable networks are more effectively
                 served by protocols that take advantage of those
                 characteristics. Peyravian et al., in their paper,
                 present their work on the new Rapid Transport Protocol,
                 designed to provide fast and effective distributed
                 network control for point-to-point and multicast
                 communications and in use as part of NBBS.
                 \par
                 Immanual et al. describe the impact of the many current
                 network protocols on the ability of networks to operate
                 effectively. In NBBS the solution to this technical
                 issue resides in the access services of the NBBS
                 transport functions, using the model of access agents
                 that exploit NBBS features to effectively deal with
                 multiple network protocols, while shielding users from
                 detailed knowledge of how this is accomplished. The
                 authors demonstrate their approach through examples of
                 access agents.
                 \par
                 Onvural et al. present two challenges to the effective
                 utilization of the ATM architecture and its
                 technologies: the development of network control
                 services meeting ATM standards without performance
                 degradation, and efficient implementation of the ATM
                 standards themselves. The authors give the current
                 status of the standards, discuss the NBBS response, and
                 describe efficient implementations of the standards in
                 NBBS.
                 \par
                 In order to control and limit the network complexity
                 visible at each node of an NBBS network, the notion of
                 an NBBS access node has been developed. Unlike an
                 intermediate node, an access node can only operate at
                 the edge of the network map and can therefore
                 effectively minimize its knowledge and support of the
                 rest of the network, without detracting from the size
                 or effectiveness of the network as a whole. Budhiraja
                 et al. present the NBBS access node and describe its
                 architecture and function.
                 \par
                 High-speed switching is a critical part of broadcast
                 networking. Lebizay et al. describe the architecture
                 and technology of the IBM transport network node and
                 explain how it supports the NBBS architecture. The
                 transport node platform uses switching technology
                 designed to avoid congestion and prevent loss of
                 data. The technology used for the switch's input and
                 output modular adapters allows them to be configured
                 for port, trunk, server, or control functions. The
                 authors show the effectiveness of the technology by
                 showing hardware and software performance figures.
                 \par
                 Modern networks necessarily contain extensive
                 capabilities for network management, and NBBS is no
                 exception. The paper by Owen describes the overall NBBS
                 approach and a number of specific functions in the
                 course of discussing such concerns as network
                 operation, use, accounting, topology, performance,
                 names, and fault management. Owen's work incorporates
                 the Common Management Information Protocol (CMIP) and
                 the Simple Network Management Protocol (SNMP)
                 standards.
                 \par
                 The next issue of the Journal will present papers on
                 medical imaging, IBM's Information FrameWork, Resource
                 Access Control Facility (RACF*), and other topics.
                 \par
                 Gene F. Hoffnagle Editor",
  acknowledgement = ack-nhfb,
}

Related entries