Table Of ContentHANDBOOK OF
NETWORKING & CONNECTIVITY
Edited by
Gary R. McClain
AP PROFESSIONAL
Boston San Diego New York
London Sydney Tokyo Toronto
This book is printed on acid-free paper. ©
Foreword and Compilation copyright © 1994 by Academic Press, Inc
"Open Systems Interconnection" copyright © 1994 by Kenneth M. Zemrowski
"Guide to Digital Communications" copyright © 1994 by E. Scott Barielle
"Implementing TCP/IP in an SNA Environment" copyright © 1994 by Shaun E. Deedy
"Integrated Services Digital Network" copyright © 1994 by Matthew N.O. Sadiku
"Metropolitan Area Networks" copyright © 1994 by Matthew N.O. Sadiku
"An Overview of the Synchronous Optical Network" copyright © 1994 by Thomas A. Maufer
and Gary C. Kessler
"X.25 and Worldwide Networking" copyright © 1994 by Fred M. Burg
"An Overview of Switched Multimegabit Data Service" copyright © 1994 by Gary C. Kessler
"Ethernet/802.3 and Token-Ring/802.5" copyright © 1994 by John Enck
"Local Area Network Cabling Considerations" copyright © 1994 by William Carltock
"Disaster Recovery Planning for Local Area Networks" copyright © 1994 by Marc Farley
"Selecting a Tape Backup System" copyright © 1994 by Nick Blozan
"Electronic Messaging" copyright © 1994 by Mikael Edholm
"Workgroup Applications" copyright © 1994 by Conall Ryan
All rights reserved.
No part of this publication may be reproduced or
transmitted in any form or by any means, electronic
or mechanical, including photocopy, recording, or
any information storage and retrieval system, without
permission in writing from the publisher.
All brand names and product names mentioned in this book are trademarks
or registered trademarks of their respective copyright owners.
AP PROFESSIONAL
955 Massachusetts Avenue, Cambridge, MA 02139
An imprint of ACADEMIC PRESS, INC.
A Division of HARCOURT BRACE & COMPANY
United Kingdom Edition published by
ACADEMIC PRESS LIMITED
24-28 Oval Road, London NW1 7DX
ISBN 0-12-482080-8
Printed in the United States of America
94 95 96 97 98 ML 9 8 7 6 5 4 3 2 1
Foreword
The content of the Handbook of Networking and Connectivity illustrates the
diversity of issues and considerations in this important area of information
management. I am grateful to the contributors for taking time out from
their hectic schedules to expand the body of knowledge in this subject area
and provide valuable reference materials for technical professionals. The
chapter contributors are all experts in their own right.
xiii
Contributors
Numbers in parentheses indicate the page on which each author's contribu-
tion begins.
E. Scott Barielle (61), Irv Shapiro and Associates, Ltd.
Nick Blozan (333), M.T. Network Solutions, Inc.
Fred M. Burg (167), AT8cJ Bell Laboratories
William Carltock (297), Equinox Systems Inc.
Shaun E. Deedy (73), LXE, Inc.
Mikael Edholm (351), Ericsson Hewlett-Packard Telecommunications, A.B.
John Enck (265), Forest Computer Inc.
Marc Farley (317), Palindrome Corporation
Gary C. Kessler (139, 231), Hill Associates
Thomas A. Maufer (139), NASA Goddard Space Flight Center
Conall Ryan (375), ON Technology Corporation
Matthew N.O. Sadiku (123, 215), Temple University
Kenneth M. Zemrowski (3), TRW Corporation
xiv
Introduction
The new era in information management during the 1990s will focus on
connectivity, with the technical staff increasingly placed in the role of not
only making sure that the hardware and software resources work, but work
together as a unified whole. Attaining this level of connectivity will require an
extensive knowledge of both proprietary and international standards as well
as an understanding of the technical considerations inherent in choosing
one networking strategy over another.
As organizations streamline to compete in the global economy, it is critical
that diverse information resources be coordinated and distributed across
departmental, national, and even international boundaries. Hardware and
software vendors have individually moved toward network management,
through local and wide area networks that form a basis for enterprise-wide
information sharing.
Network management professionals will be key players in IS departments
of the 1990s. Many vendors are building some level of standards compliance
into their product offerings. However, the IS stafiF will be responsible for
understanding the nuances of this compliance, discerning the levels of com-
pliance and approach that is correct for their organizations, and making the
strategy work.
The overall purpose of the Handbook of Networking and Connectivity is to
serve as a comprehensive networking and connectivity reference book for
MIS networking and systems professionals. The reader will be able to look to
this book as a source for understanding the current connectivity standards
in use, as well as hardware and software options, and as a guide for solving
specific problems that arise in designing and maintaining organizational
networks.
xv
1
Open Systems
Interconnection
Kenneth M. Zemrowskl CDP, CCP
TRW Corporation
PURPOSE OF OSI
The aim of Open Systems Interconnection (OSI) is to allow, with a mini-
mum of technical agreement outside the interconnection standards, the
interconnection of computer systems:
• from different manufacturers,
• under different management,
• of different levels of complexity, and
• of different technologies.
The OSI standards were developed to allow computer systems built by
different vendors to exchange data; the previous use of proprietary network
architectures hampered the development of large, multivendor networks.
Even though these computer systems have different operating systems and
vary in how data is processed internally, as long as the information that
passes between the processors conforms to the OSI international standards,
information can be interpreted upon receipt and communication is possi-
ble. Openness, in this context, does not necessarily imply that the network is
public.
With the growing importance of computer interconnection, there was a
need for increasingly sophisticated and powerful communications stand-
3
4 Handbook of Networking and Connectivity
ards, as well as improved planning to support future growth and produce
well-coordinated, timely standards. By the early 1980s, proprietary solutions
were causing difficulty in establishing networks, as networks grew more
complicated and there was increased demand for multivendor systems. By
the 1990s, users demanded open system architectures that would permit the
use of new components in a system, whether for purposes of technology
insertion or simply to enhance their choice of vendors.
By developing a Reference Model of OSI, the International Organization
for Standardization (ISO) expected to provide a common basis for the
coordination of standards development for the purpose of systems intercon-
nection, while allowing the existing standards to be placed into perspective
within the overall reference model. This was an attempt to provide improved
capabilities in the then unstandardized area of process-to-process communi-
cations (in contrast to node-to-node) while preserving the investment in
existing protocols.
OSI has reached a point where there is arguably a critical mass of software
available to begin fulfilling the promises of OSI. There are now production
quality implementations for a variety of platforms. Vendors cooperate to
verify interoperability, and there are also independent agencies that certify
conformance to protocol standards and test interoperability of implementa-
tions (see the section on Conformance Testing).
PRINCIPLES OF OSI
OSI involves the exchange of information between open systems, but not the
internal functioning of each individual system. (Open systems environments
are concerned with the openness of individual systems and the portability of
the applications. OSI is primarily concerned with interoperability.) This
exchange of information is facilitated by the mutual use of applicable stand-
ards and agreement on the use of options and parameters from those stand-
ards.
The OSI Reference Model does not provide a protocol (or even selection
of protocols). Rather, it provides a conceptual and functional framework
that allows international teams of experts to work productively and inde-
pendently on the development of standards for each layer of OSI. Another
consideration was the ability to accommodate existing protocols in the
framework. The OSI model has sufficient flexibility to accommodate ad-
vances in technology and expansion in user demands. This flexibility is also
intended to allow the phased transition from existing implementations to
Open Systems Interconnection 5
OSI standards. Besides flexibility to accommodate change, the framework
helps support planning.
OSI addresses both the network interconnections and the interworking of
applications. Applications may be user-developed or include standard appli-
cations such as message handling (using X.400, MHS) or file transfer (file
transfer, access, and management [FTAM] ). In addition, OSI standards now
explicitly address conformance testing, which forms an important step in
assuring interoperability.
The important principles are that (1) each layer performs a unique, ge-
neric, well-defined function, and (2) layer boundaries are designed so that
the amount of information flowing between any two adjacent layers is mini-
mized. A particular layer has to provide a sufficient number of services to the
layer immediately above for the latter to perform its functions.
The functions of each of the protocol layers will be explained later in this
chapter. The protocols can be connection-oriented or connectionless. In
connection-oriented protocols, a user must set up a virtual connection,
which is valid for the life of the communications activity, and disappears
when the communications activity disappears. The converse of this is con-
nectionless activity, whereby the user does not set up a virtual connection
but communicates by transmitting individual "pieces" of information. An
example of the former is a telephone conversation; an example of the latter
is message delivery by the postal service.
OSI standards have been developed under the aegis of the ISO and Inter-
national Electrotechnical Committee (IEC), with many of the standards
developed collaboratively with the International Telephone and Telegraph
Consultative Committee (CCITT). Where the CCITT and JTC1 (ISO/IEC
Joint Technical Committee 1) have equivalent specifications, the documents
are technically aligned.
BASIC PRINCIPLES OF LAYERING
The first step in OSI standards development was the creation of an OSI
Reference Model. This model is divided into seven layers; each layer pro-
vides a well-defined set of functions necessary for the effective transmission
of data. Each of these layers provides a service to the layer above by carrying
on a conversation with the same layer on another processor. The rules and
conventions ofthat conversation are called a protocol. The information that
is passed between a layer on one processor and the corresponding layer (or
peer entity) on another processor is called a protocol data unit (PDU).
6 Handbook of Networking and Connectivity
The OSI framework imposes a seven layer model to define the communi-
cations processes that occur between systems. The Basic Reference Model
(ISO 7498-1:1984) describes seven layers, as well as the relationships be-
tween a layer and the layer below it. The reference model, and other stand-
ards, define the services necessary to perform the functions. Other standards
specify the protocols for how the information is actually communicated.
Only protocols are actually implemented; the service definitions describe
the structures and functions that are performed, while the protocols de-
scribe the detailed formats and dialogs.
Service primitives are special messages that define the services that a layer
provides. The details of how the services are implemented are transparent to
the service user, which is usually the next upper layer. Communication
between layers is via a service access point (SAP), which is a special location
through which service primitives pass. Service request and service response
information passes between adjacent layers at the service access point. Serv-
ice definitions are defined primarily for the purpose of developing protocol
specifications.
For any layer (N+l), the reference model describes services provided by
the next lower layer (N). For example, the Application Layer (Layer 7) uses
services provided by the Presentation Layer (Layer 6).
Layers 1 to 6, together with the physical media for OSI, provide a step-by-
step enhancement of communication services. The boundary between two
layers identifies a stage in this enhancement of services at which an OSI
service standard is defined, while the functioning of the layers is governed by
OSI protocol standards. In some cases, there are several standards defining
services and protocols for a particular layer, especially at the Application
Layer.
According to ISO/IEC 7498-1, several principles are used in determining
the seven layers in the Reference Model and will also guide future decisions.
1. Do not create so many layers as to make the system engineering task of
describing and integrating the layers more difficult than necessary.
2. Create a boundary at a point where the description of services can be
small and the number of interactions across the boundary are mini-
mized.
3. Create separate layers to handle functions that are manifestly differ-
ent in the process performed or the technology involved.
4. Collect similar functions into the same layer.
Open Systems Interconnection 7
5. Select boundaries at a point that past experience has demonstrated to
be successful.
6. Create a layer of easily localized functions so that the layer could be
totally redesigned and its protocols changed in a major way to take
advantage of new advances in architectural, hardware, or software
technology without changing the services expected from and pro-
vided to the adjacent layers.
7. Create a boundary where it may be useful at some point in time to
have the corresponding interface standardized.
Within the abstract model, an open system is logically composed of an
ordered set of subsystems. Adjacent subsystems communicate through their
common boundary. Entities in the same layer are termed peer entities.
It is important to note the distinction between the abstract model, which
is expressed in ISO 7498-1, and the real model, which is embodied in imple-
mentations. The real model does not require discrete separations between
the layers, which may reduce the protocol overhead. However, a "glued
together" stack might make it more difficult to perform conformance tests,
which rely on an exposed stack at various layers. Typically, there will be an
exposed interface to the Network Layer, which is necessary if a node is to
serve as an intermediate system, and to the Transport Layer. The upper
layers are often grouped together without having explicit separations be-
tween the software implementing each layer. (See Figure 1.1.)
Data Flow {Peer-to-Peer}
Application Application
Presentation Presentation
Session Session
Transport Transport
p
Network Network Network Network
Data Link Data Link Data Link Data Link
Physical Physical Physical Physical
Figure 1.1 Data flow and control flow from end system through intermediate nodes
to end system.
