An Introduction to ATM Networks 1st Edition Harry G. Perros

Visit https://ebookfinal.com to download the full version and
explore more ebooks
An Introduction to ATM Networks 1st Edition
Harry G. Perros
_____ Click the link below to download _____
https://ebookfinal.com/download/an-introduction-to-atm-
networks-1st-edition-harry-g-perros/
Explore and download more ebooks at ebookfinal.com

Here are some suggested products you might be interested in.
Click the link to download
ATM Networks Concepts and Protocols 2nd Edition Sumit
Kasera
https://ebookfinal.com/download/atm-networks-concepts-and-
protocols-2nd-edition-sumit-kasera/
Connections An Introduction to the Economics of Networks
Sanjeev Goyal
https://ebookfinal.com/download/connections-an-introduction-to-the-
economics-of-networks-sanjeev-goyal/
Social Capital An Introduction to Managing Networks
Kenneth W. Koput
https://ebookfinal.com/download/social-capital-an-introduction-to-
managing-networks-kenneth-w-koput/
Introduction to Logic 3rd Edition Harry J. Gensler
https://ebookfinal.com/download/introduction-to-logic-3rd-edition-
harry-j-gensler/

From GSM to LTE Advanced An Introduction to Mobile
Networks and Mobile Broadband Martin Sauter
https://ebookfinal.com/download/from-gsm-to-lte-advanced-an-
introduction-to-mobile-networks-and-mobile-broadband-martin-sauter/
An Introduction to Petroleum Technology Economics and
Politics 1st Edition James G. Speight
https://ebookfinal.com/download/an-introduction-to-petroleum-
technology-economics-and-politics-1st-edition-james-g-speight/
Political Philosophy An Introduction 1st Edition Richard
G. Stevens
https://ebookfinal.com/download/political-philosophy-an-
introduction-1st-edition-richard-g-stevens/
Stories of Culture and Place An Introduction to
Anthropology 1st Edition Michael G. Kenny
https://ebookfinal.com/download/stories-of-culture-and-place-an-
introduction-to-anthropology-1st-edition-michael-g-kenny/
An introduction to statistical concepts for education and
behavioral sciences 1st Edition Richard G. Lomax
https://ebookfinal.com/download/an-introduction-to-statistical-
concepts-for-education-and-behavioral-sciences-1st-edition-richard-g-
lomax/

An Introduction to ATM Networks 1st Edition Harry G.
Perros Digital Instant Download
Author(s): Harry G. Perros
ISBN(s): 9780471498278, 0471498270
Edition: 1
File Details: PDF, 19.22 MB
Year: 2001
Language: english

An Introduction to ATM Networks

This page intentionally left blank

An Introduction to ATM
Networks
Harry G. Perros
NC State University, Raleigh, USA
JOHN WILEY & SONS, LTD
Chichester • New York • Weinheim • Brisbane • Singapore • Toronto

Copyright © 2002 John Wiley & Sons, Ltd
Baffins Lane, Chichester,
West Sussex PO19 1UD, England
National 01243 779777
International (+44) 1243 779777
e-mail (for orders and customer service enquiries): cs-books@wiley.co.uk
Visit our Home Page on http://www.wiley.co.uk or http://www.wiley.com
All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted, in any form or by any means, electronic, mechanical, photocopying,
recording, scanning or otherwise, except under the terms of the Copyright, Designs and
Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency
Ltd, 90 Tottenham Court Road, London W1P OLP, UK without the permission of the
Publisher and the copyright owner, with the exception of any material supplied specifi-
cally for the purpose of being entered and executed on a computer system, for the exclusive
use by the purchaser of the publication.
Other Wiley Editorial Offices
John Wiley & Sons, Inc., 605 Third Avenue,
New York, NY 10158-0012, USA
Wiley-VCH Verlag GmbH
Pappelallee 3, D-69469 Weinheim, Germany
John Wiley & Sons Australia, Ltd
33 Park Road, Milton, Queensland 4064, Australia
John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01,
Jin Xing Distripark, Singapore 129809
John Wiley & Sons (Canada) Ltd, 22 Worcester Road,
Rexdale, Ontario, M9W 1L1, Canada
Library of Congress Cataloging-in-Pubttcation Data
Perros, Harry G.
An introduction to ATM networks / Harry G. Perros.
p. cm.
Includes bibliographical references and index.
ISBN 0-471-49827-0 (alk. paper)
1. Asynchronous transfer mode. I. Title.
TK5105.35.P48 2001
004. 6'6—dc21 2001026646
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
ISBN 0-471-49827-0
Typeset in 10/12pt Times Roman by Laser Words, Chennai, India.
Printed and bound in Great Britain by Biddies Ltd, Guildford Surrey.
This book is printed on acid-free paper responsibly manufactured from sustainable
forestry in which at least two trees are planted for each one used for paper production.

About the Author
Harry G. Perros received the BSc degree in mathematics in 1970 from Athens University,
Greece, the MSc degree in operational research with computing from Leeds University,
England, in 1971, and the PhD degree in operations research from Trinity College Dublin,
Ireland, in 1975.
From 1976 to 1982 he was an Assistant Professor in the Department of Quantitative
Methods, University of Illinois at Chicago. In 1982 he joined the Department of Computer
Science, North Carolina State University, as an Associate Professor, and since 1988 he has
been a Professor. He has spent sabbaticals at INRIA, Rocquencourt, France, University
of Paris 6, France, and NORTEL, Research Triangle Park, North Carolina.
He has published extensively in the area of performance modeling of computer and
communication systems, and has organized several national and international conferences.
He has also published a monograph entitled Queueing Networks with Blocking: Exact and
Approximate Solutions (Oxford University Press). He is the chairman of the IFIP Working
Group 6.3 on the Performance of Communication Systems. In his free time, he likes to
sail on board the Aegean, a Pearson 31!

Contents
Preface xi
List of Abbreviations xv
Part 1: Introduction and Background 1
1 Introduction 3
1.1 The Asynchronous Transfer Mode 3
1.2 Standards Committees 4
Problems 9
2 Basic Concepts from Computer Networks 11
2.1 Communication Networking Techniques 11
2.2 The Open System Interconnection (OSI) Reference Model 13
2.3 Data Link Layer 14
2.4 The High Data Link Control (HDLC) Protocol 18
2.5 Synchronous Time Division Multiplexing (TDM) 20
2.6 The Logical Link Control (LLC) Layer 22
2.7 Network Access Protocol X.25 24
2.8 The Internet Protocol (IP) 26
2.8.1 The IP Header 26
2.8.2 IP Addresses 28
2.8.3 ARP, RARP and ICMP 30
2.8.4 IP Version 6 (IPv6) 31
Problems 31
3 Frame Relay 33
3.1 Motivation and Basic Features 33
3.2 The Frame Relay UNI 35
3.3 Congestion Control 38
Problems 41
Part 2: The ATM Architecture 43
4 Main Features of ATM Networks 45

viii CONTENTS
4.1 Introduction 45
4.2 Structure of the ATM Cell Header 48
4.2.1 Generic Flow Control (GFC) 48
4.2.2 Virtual Path Identifier/Virtual Channel Identifier (VPI/VCI) 48
4.2.3 Payload Type Indicator (PTI) 50
4.2.4 Cell Loss Priority (CLP) Bit 51
4.2.5 Header Error Control (HEC) 51
4.3 The ATM Protocol Stack 52
4.3.1 The Physical Layer 52
4.3.2 The ATM Layer 53
4.3.3 The ATM Adaptation Layer 55
4.3.4 Higher Level Layers 56
4.4 ATM Interfaces 56
4.5 The Physical Layer 58
4.5.1 The Transmission Convergence (TC) Sublayer 58
4.5.2 The Physical Medium-Dependent (PMD) Sublayer 59
4.5.3 ATM Physical Layer Interfaces 60
4.6 UTOPIA and WIRE 64
Problems 64
5 The ATM Adaptation Layer 67
5.1 Introduction 67
5.2 ATM Adaptation Layer 1 (AAL 1) 69
5.2.1 The AAL 1 SAR sublayer 69
5.2.2 The AAL 1 CS sublayer 71
5.4 ATM Adaptation Layer 3/4 (AAL 3/4) 76
Problems 80
6 ATM Switch Architectures 81
6. 1 Introduction 81
6.2 Space-Division Switch Architectures 83
6.2.1 The Cross-Bar Switch 83
6.2.2 Banyan Networks 86
6.2.3 Clos Networks 93
6.2.4 Switch Architectures with N2
Disjoint Paths 93
6.3 Shared Memory ATM Switch Architectures 94
6.4 Shared Medium ATM Switch Architectures 96
6.5 Nonblocking Switches with Output Buffering 98
6.6 Multicasting in an ATM Switch 99
6.7 Scheduling Algorithms 100
6.8 The Lucent AC120 Switch 103
6.9 Performance Evaluation of an ATM Switch 105
Problems 106
Appendix: A Simulation Model of an ATM Multiplexer—Part 1 107

CONTENTS ix
7 Congestion Control in ATM Networks 111
7. 1 Traffic Characterization 111
7. 1.1 Standardized Traffic Descriptors 114
7. 1.2 Empirical Models 114
7. 1.3 Probabilistic Models 115
7.2 Quality of Service (QoS) Parameters 117
7.3 ATM Service Categories 120
7.4 Congestion Control 122
7.5 Preventive Congestion Control 122
7.6 Call Admission Control (CAC) 123
7.6. 1 Equivalent Bandwidth 125
7.6.2 The ATM Block Transfer (ABT) Scheme 128
7.6.3 Virtual Path Connections 129
7.7 Bandwidth Enforcement 131
7.7. 1 The Generic Cell Rate Algorithm (GCRA) 132
7.7.2 Packet Discard Schemes 135
7.8 Reactive Congestion Control 136
7.8.1 The Available Bit Rate (ABR) Service 136
Problems 141
Appendix: A Simulation Model of an ATM Multiplexer—Part 2 142
Appendix: Estimating the ATM Traffic Parameters of a Video Source 144
Part 3: Deployment of ATM 147
8 Transporting IP Traffic Over ATM 149
8.2 LAN Emulation (LE) 150
8.3 Classical IP and ARP over ATM 154
8.3. 1 ATMARP 155
8.3.2 IP Multicasting over ATM 156
8.4 Next Hop Resolution Protocol (NHRP) 160
8.5 IP Switching 163
8.6 Tag Switching 166
8.7 Multi-Protocol Label Switching (MPLS) 172
Problems 174
9 ADSL-Based Access Networks 175
9.2 The ADSL Technology 178
9.2. 1 The Discrete Multi-Tone (DMT) Technique 180
9.2.2 Bearer Channels 181
9.2.3 The ADSL Super Frame 182
9.3 Schemes for Accessing Network Service Providers 182
9.3.1 The L2TP Access Aggregation Scheme 184
9.3.2 The PPP Terminated Aggregation Scheme 185
Problems 186

x CONTENTS
Part 4: Signaling in ATM Networks 187
10 Signaling over the UNI 189
10.1 Connection Types 189
10.2 The Signaling Protocol Stack 190
10.3 The Signaling ATM Adaptation Layer (SAAL) 190
10.3.1 The SSCOP 191
10.3.2 Primitives 192
10.4 The Signaling Channel 194
10.5 ATM Addressing 195
10.6 The Format of the Signaling Message 197
10.7 The Signaling Protocol Q.2931 199
10.7. 1 Information Elements (IE) 199
10.7.2 Q.2931 Messages 202
10.8 The Signaling Protocol Q.2971 204
10.9 Leaf Initiated Join (LIJ) Capability 206
10.10 ATM Anycast Capability 208
Problems 209
11 The Private Network-Network Interface (PNNI) 211
11.1 Introduction 211
11.2 The PNNI Routing Protocol 212
11.2.1 The Lowest-Level Peer Groups 212
11.2.2 The Next Level of Peer Groups 214
11.2.3 Uplinks 215
11.2.4 Information Exchange in the PNNI Hierarchy 216
11.2.5 The Highest-Level Peer Group 217
11.2.6 A Node's View of the PNNI Hierarchy 219
11.2.7 Address Summarization 220
11.2.8 Level Indicators 222
11.2.9 Path Selection 222
11.3 The PNNI Signaling Protocol 223
Problems 224
Appendix: List of standards 227
Index 229

Preface
ATM networks was the subject of intense research and development from the late 1980s
to the late 1990s. Currently, ATM is a mature networking technology and is regularly
taught in universities and in short professional courses. This book was written with a
view to be used as a textbook in a second course on computer networks at the graduate
level or senior undergraduate level. Also, it was written for networking engineers out in
the field who would like to learn more about ATM networks. A prerequisite for this book
is basic knowledge of computer networking principles.
The book is organized into the following parts:
Part One: Introduction and Background
Part Two: The ATM Architecture
Part Three: Deployment of ATM
Part Four: Signaling in ATM Networks.
Part One, 'Introduction and Background', contains a variety of topics which are part
of the background necessary for understanding the material in this book. It consists of
Chapters 1, 2 and 3. Chapter 1 contains a discussion of what caused the development of
ATM networks, and a brief description of the various standards committees that feature
prominently in the development of ATM networks. Chapter 2 gives a review of basic
concepts of computer networks that are used in this book. This chapter can be skipped by
the knowledgeable reader. Chapter 3 is dedicated to frame relay, where we describe the
motivation behind the development of frame relay and its basic features, the frame relay
UNI, and congestion control. It is educationally constructive to understand how frame
relay works, since it is a very popular networking solution and it has many common
features with ATM networks, such as layer two switching, no error or flow control between
two adjacent nodes, and similar congestion control schemes.
Part Two, 'The ATM Architecture', focuses on the main components of the ATM
architecture. It consists of Chapters 4, 5, 6 and 7. In Chapter 4, the main features of
the ATM architecture are presented. An ATM packet, known as a cell, has a fixed
size and it is equal to 53 bytes. We start with a brief account of the considerations
that led to the decision to use such a small packet. Then, we describe the structure of
the header of the ATM cell, the ATM protocol stack, and the various ATM interfaces.
We conclude this chapter with a description of the physical layer that supports ATM
networks, and the various public and private interfaces. In Chapter 5, we describe the
ATM adaptation layer. The purpose of this layer is to isolate higher protocol layers and
applications from the specific characteristics of ATM. Four different ATM adaptation

xii PREFACE
layers are described, namely ATM adaptation layers 1, 2, 3/4 and 5. Chapter 6 is dedi-
cated to ATM switch architectures, and the following different classes of architecture are
presented: space-division switches, shared memory switches, and shared mediumswitches.
We describe various architectures that have been proposed within each of these three
classes. Also, to give the reader a feel of a real-life switch, the architecture of a commer-
cial switch is described. We conclude this chapter by describing various algorithms for
scheduling the transmission of cells out of an output port of an ATM switch. Finally,
Chapter 7 deals with the interesting problem of congestion control in ATM networks.
We first present the various parameters used to characterize ATM traffic, the various
Quality of Service (QoS) parameters, and the standardized ATM classes. In the rest of the
chapter, we focus on the two classes of congestion control schemes, namely, preven-
tive and reactive congestion control. We introduce the preventive congestion control
scheme, and present various call admission control algorithms, the GCRA bandwidth
enforcement algorithm, and cell discard policies. Finally, we present the Available Bit
Rate (ABR) scheme, a reactive congestion control scheme standardized by the ATM
Forum.
Part Three, 'Deployment of ATM', deals with the different topics: how IP traffic is
transported over ATM, and ADSL-based access networks. In Chapter 8, we describe
various schemes used to transport IP traffic over ATM. We first present ATM Forum's
LAN Emulation (LE), a solution that enables existing LAN applications to run over an
ATM network. Then, we describe the lETF's classical IP and ARP over ATM and Next
Hop Resolution Protocol (NHRP) schemes, designed for carrying IP packets over ATM.
The rest of the chapter is dedicated to three techniques, IP switching, tag switching,
and Multi-Protocol Label Switching (MPLS). IP switching inspired the development of
tag switching, which at the moment is being standardized by IETF under the name of
multi-protocol label switching. Chapter 9 is dedicated to Asymmetric Digital Subscriber
Line (ADSL) technology, which can be used in residential access networks to provide
basic telephone services and access to the Internet. We describe the Discrete Multi-Tone
(DMT) technique used to transmit the information over the telephone twisted pair, the
seven bearer channels, the fast and interleaved paths, and the ADSL super frame. Finally,
we discuss architectures for accessing network service providers.
Part Four, 'Signaling in ATM Networks', focuses on the signalingprotocols used to set-
up a Switched Virtual Connection (SVC). In Chapter 10,we review the signaling protocols
used to establish a point-to-point connection and a point-to-multipoint connection over
the private UNI. The signaling protocol for establishing a point-to-point connection is
described in ITU-T's Q.2931 standard, and the signaling protocol for establishing a point-
to-multipoint connection is described in ITU-T's Q.2971 standard. We first describe a
specialized ATM adaptation layer, known as the signaling AAL (SAAL), which is used
by both protocols. Then, we discuss in detail the signaling messages and procedures used
by Q.2931 and Q.2971. In Chapter 11, we examine the Private Network-NetworkInterface
(PNNI) used to route a new call from an originating UNI to a destination UNI. PNNI
consists of the PNNI routing protocol and the PNNI signaling protocol. We first describe
the PNNI routing protocol in detail, and then we briefly discuss the PNNI signaling
protocol.
At the end of each chapter there are some problems given. Also, in Chapters 6 and 7
there are three simulation projects, designed to help the reader understand better some of
the intricacies of ATM networks.

To develop a deeper understanding of ATM networks, one has to dig into the various
documents produced by the standards bodies. Most of these documents are actually very
readable! A list of standards which are relevant to the material presented here can be
found at the end of the book.
Finally, in the ATM networks field there is an abundance of abbreviations, and the reader
is strongly encouraged to learn some of them. When in doubt, the list of abbreviations
given may be of help!
Harry Perros
xiii

List ofAbbreviations
AAL ATM adaptation layer
ABR available bit rate
ABT ATM block transfer
ACR allowable cell rate
ADSL asymmetric digital subscriber line
AFI authority and format identifier
ANP AAL 2 negotiation procedure
APON ATM passive optical networks
ARP address resolution protocol
ARQ automatic repeat request
ATM asynchronous transfer mode
ATU-C ADSL transceiver unit at the central office
ATU-R ADSL transceiver unit at the remote terminal
BAS broadband access server
BCOB-A broadband connection oriented bearer class A
BCOB-C broadband connection oriented bearer class C
BCOB-X broadband connection oriented bearer class X
B-frame bi-directional-coded frame
B-ICI broadband inter-carrier interface
BECN backward explicit congestion notification
BGP border gateway protocol
BOM beginning of message
BT burst tolerance
BUS broadcast and unknown server
CAC call admission control
CBR constant bit rate
CCITT International Telegraph and Telephone Consultative Committee
CCR current cell rate
CDVT cell delay variation tolerance
CER cell error rate
CI connection identifier
CIDR classless inter-domain routing
CIR committed information rate
CLEC competitive local exchange carrier
CLLM consolidated link layer management

XVi LIST OFABBREVIATIONS
CLNAP connectionless network access protocol
CLNIP connectionless network interface protocol
CLP cell loss priority bit
CLR cell loss rate
CLS connectionless server
CMR cell misintertion rate
CO central office
COM continuation of message
CoS class of service
CPS common part sublayer
CRC cyclic redundant check
CR-LDP constraint routing-label distribution protocol
CS convergence sublayer
CTD cell transfer delay
DBR deterministic bit rate
DCC data country code
DCE data communication equipment
DMCR desirable minimum cell rate
DMT discrete multi-tone
DOCSIS data-over-cable service interim specification
DSL digital subscriber loop
DSLAM ADSL access multiplexer
DSP domain-specific part
DTE data terminal equipment
DTL designated transit list
EFCN explicit forward congestion notification
EOM end of message
ER explicit rate
ESI end system identifier
FCS frame check sequence
FDM frequency division multiplexing
EEC forwarding equivalent class
FECN forward explicit congestion notification
FIB forwarding information base
FRAD frame relay access devices
FRP/DT fast reservation protocol with delayed transmission
FTTB fiber to the basement
FTTC fiber to the curb
FTTCab fiber to the cabinet
FTTH fiber to the home
GCRA generic cell rate algorithm
GFR guaranteed frame rate
GSMP general switch management protocol
HDLC high-level data link control
HDSL high data rate DSL
HEC header error control
HFC hybrid fiber coaxial

HO-DSP high-order DSP
IBP interrupted Bernoulli process
ICD international code designator
ICMP internet control message protocol
IDI initial domain identifier
IDP initial domain part
IDSL ISDN DSL
IE information elements
IFP interrupted fluid process
IFMP Ipsilon's flow management protocol
I-frame intra-coded frame
IGMP internet group management protocol
IISP interim interswitch signaling protocol
InATMARP inverse ATMARP
ILEC incumbent local exchange carrier
IP internet protocol
IPP interrupted Poisson process
ISO International Organization of Standards
ISP Internet service provider
ITU International Telecommunication Union
IWU interworking unit
L2TP layer 2 tunnel protocol
LAC L2TP access concentrator
LDP label distribution protocol
LE LAN emulation
LE-ARP LAN emulation address resolution
LECID LE client identifier
LER label edge router
LIS logical IP subnet
LIJ leaf initiated join
LMDS local multipoint distribution services
LMI local management interface
LSP label switched path
LSR label switching router
LUNI LAN emulation user to network interface
MARS multicast address resolution server
MBS maximum burst size
MCR minimum cell rate
MCS multicast servers
ME mapping entity
MFS maximum frame size
MMBP Markov modulated Bernoulli process
MMPP Markov modulated Poisson process
MPLS multi-protocol label switching
MPOA multi-protocol over ATM
MTU maximum transfer unit
NAS network access server
xvii

XViii LIST OF ABBREVIATIONS
NBMA non broadcast multiaccess network
NHC next hop client
NHRP next hop resolution protocol
NHS next hop server
NNI network node interface
NRT-VBR non-real-time variable bit rate
NRT-SBR non-real-time statistical bit rate
NSAP network service access point
NSP network service provider
NTR network timing reference
OC optical carrier
OLT optical line terminator
ONU optical network unit
OSI open system interconnection reference model
OSPF open shortest path first
PCM pulse code modulation
PCR peak cell rate
PDH plesiochronous digital hierarchy
PDU protocol data unit
P-frame predictive-coded frame
PGL peer group leader
PIM protocol independent multicast
PMD physical medium dependent sublayer
PNNI private network-network interface or private network node interface
PON passive optical network
PPP point-to-point protocol
PTI payload type Indicator
PTSE PNNI topology state element
PTSP PNNI topology state packet
PVC permanent virtual connection
QAM quadrature amplitude modulation
RADIUS remote authentication dial in user service
RCC routing control channel
RM resource management
ROC regional operations center
RSVP resource reservation protocol
RT-VBR real-time variable bit rate
RT-SBR real-time statistical bit rate
SAAL signaling AAL
SAR segmentation-and-reassembly sublayer
SBR statistical bit rate
SCR sustained cell rate
SDH synchronous digital hierarchy
SDU service data unit
SDSL symmetric DSL
SEL selector
SMDS switched multimegabit data service

SONET synchronous optical network
SSCF service-specific connection function
SSCOP service-specific connection oriented protocol
SSCS service specific convergence sublayer
SSM single segment message
STF start field
STM synchronous transfer mode
STS-1 synchronous transport signal level 1
SVC switched virtual connection
TC transmission convergence sublayer
TOP tag distribution protocol
TER tag edge router
TFIB tag forwarding information base
TSR tag switching router
TTL time to live
UBR unspecified bit rate
UNI user network interface
VCC virtual channel connection
VCI virtual channel identifier
VDSL very high data rate DSL
VPI virtual path identifier
WDM wavelength division multiplexing
xDSL x
-type digital subscriber line
xix

Part 1
Introduction and Background

1
Introduction
In this chapter, we introduce the Asynchronous Transfer Mode (ATM) networking tech-
nique, and discuss the forces that gave rise to it. Then, we describe some of the well
known national and international standards committees involved with the standardization
process of networking equipment.
1.1 THE ASYNCHRONOUS TRANSFER MODE
ATM is a technology that provides a single platform for the transmission of voice,
video and data at specified quality of service and at speeds varying from fractional Tl
(i.e. nX64 Kbps), to Gbps. Voice, data and video are currently transported by different
networks. Voice is transported by the public telephone network, and data by a variety
of packet-switched networks. Video is transported by networks based on coaxial cables,
satellites and radio waves, and to a limited extent, by packet-switched networks.
To understand what caused the development of ATM, we have to go back to the 1980s!
During that decade, we witnessed the development of the workstation and the evolution
of the optical fiber. A dramatic reduction in the cost of processing power and associated
peripherals, such as main memory and disk drives, led to the development of powerful
workstations capable of running large software. This was a significant improvement over
the older 'dumb terminal'. These workstations were relatively cheap to buy, easy to
install and interconnect, and they enabled the development of distributed systems. As
distributed systems became more commonplace, so did the desire to move files over the
network at a higher rate. Also, there was a growing demand for other applications, such as
videoconferencing, multimedia, medical imaging, remote processing and remote printing
of a newspaper. At the same time, optical fiber technology evolved very rapidly, and by
the end of the 1980s a lot of optical fiber had been installed. Optical fiber permitted high
bandwidth and very low bit-error rate.
These technological developments, coupled with the market needs for faster intercon-
nectivity, gave rise to various high-speed wide-area networks and services, such as frame
relay, Asynchronous Transfer Mode (ATM) and Switched Multimegabit Data Services
(SMDS).
ATM was standardized by ITU-T in 1987. It is based on packet-switching and is
connection oriented. An ATM packet, known as a cell, is a small fixed-size packet with a
payload of 48 bytes and a 5-byte header. The reason for using small packets was motivated
mostly by arguments related to the transfer of voice over ATM.
Unlike IP networks, ATM has built-in mechanisms that permit it to provide different
quality of service to different types of traffic. ATM was originally defined to run over

4 INTRODUCTION
high-speed links. For instance, in North America, the lowest envisioned speed was OC-3.
which corresponds to about 155 Mbps. It should be noted that the fastest network in
the late 1980s was the FDDI (Fiber Distributed Data Interface), which ran at 100Mbps.
However, as ATM became more widely accepted, it was also defined over slow links,
such as fractional Tl, i.e., nX64 Kbps.
In the early 1990s, ATM was poised to replace well-established local and wide area
networks such as Ethernet and IP networks. ATM was seen as a potential replacement for
Ethernet because it ran faster, and also provided a good quality of service. At that time,
Ethernet ran at 10Mbps, but due to software bottlenecks, its effective throughput was
around 2 Mbps. Also, since ATM has its own addressing system, and it can set-up and
route connections through the network, it was seen as a potential foe of IP networks. In
view of this, Ethernet and IP networks were declared by the ATM aficionados as 'dead'!
Interestingly enough, Ethernet made a dramatic come-back when it was defined to run
at 100Mbps and later on at 1Gbps. As a result, ATM lost the battle to the 'desktop',
i.e. it never became the preferred networking solution for interconnecting workstations
and personal computers at a customer's premises. Also, in the mid-1990s, we witnessed
a new wave of high-speed IP routers and a strong effort to introduce quality of service
in IP networks. As a result, one frequently hears cries that it is the ATM technology that
is now 'dead'!
ATM is a mature networking technology, and it is still the only networking technology
that provides quality of service. ATM networks are used in a variety of environments.
For instance, it is widely used in the backbone of Internet Service Providers (ISP) and
in campus networks to carry Internet traffic. ATM wide area networks have also been
deployed to provide point-to-point and point-to-multipoint video connections. Also, there
are on-going projects in telecommunication companies aiming at replacing the existing
trunks used in the telephone network with an ATM network.
On a smaller scale, ATM is used to provide circuit emulation, a service that emulates
a point-to-point T1/E1 circuit and a point-to-point fractional T1/E1 circuit over an ATM
network. ATM is the preferred solution for ADSL-based residential access networks used
to provide access to the Internet and basic telephone services over the phone line. Also,
it is used in Passive Optical Networks (PON) deployed in residential access networks.
We conclude this section by noting that arguments in favor and against existing and
emerging new networking technologies will most likely continue for a long time. There is
no argument, however, that these are indeed very exciting times as far as communication
systems are concerned!
1.2 STANDARDS COMMITTEES
Standards allow vendors to develop equipment to a common set of specifications. Providers
and end-users can also influence the standards so that the vendors' equipment conforms
to certain characteristics. As a result of the standardization process, one can purchase
equipment from different vendors without being bound to the offerings of a single vendor.
There are two types of standards, namely defacto and de jure. Defacto standards are
those which were first developed by a single vendor or a consortium, and then they were
accepted by the standards bodies. Dejure standards are those generated through consensus
within national or international standards bodies. ATM, for instance, is the result of the
latter type of standardization.

STANDARDS COMMITTEES 5
Several national and international standards bodies are involved with the standardiza-
tion process in telecommunication, such as the International Telecommunication Union
(ITU), the International Organization for Standardization (ISO), the American National
Standards Institute (ANSI), the Institute of Electrical and Electronics Engineering (IEEE),
the Internet Engineering Task Force (IETF), the ATM Forum, and the Frame Relay Forum.
The organizational structure of these standards bodies is described below.
The ITU-T and the ATM Forum are primarily responsible for the development of
standards for ATM networks. ITU-T concentrates mainly on the development of standards
for public ATM networks, whereas the ATM Forum concentrates on private networks. The
ATM Forum was created because many vendors felt that the ITU-T standardization process
was not moving fast enough, and also because there was an emerging need for standards
for private ATM networks. In general, ITU-T tends to reflect the view of network operators
and national administrations, whereas the ATM Forum tends to represent the users and
the Customer Premise Equipment (CPE) manufacturers.The two bodies compliment each
other and work together to align their standards with each other.
The International Telecommunication Union (ITU)
ITU is a United Nations specialized agency whose job is to standardize international
telecommunications. ITU consists of the following three main sections: the ITU Radio-
communications Sector (ITU-R), the ITU Telecommunications Standardization Sector
(ITU-T), and the ITU Development Sector (ITU-D).
The ITU-T's objective is telecommunications standardization on a worldwide basis.
This is achieved by studying technical, operating and traffic questions, and adopting
recommendations on them. ITU-T was created in March 1993, and it replaced the former
well-known standards committee, the International Telegraph and Telephone Consulta-
tive Committee, whose origins go back over 100 years. This committee was commonly
referred to as the CCITT, which are the initials of its name in French.
ITU-T is formed by representatives from standards organizations, service providers, and
more recently, by representatives from vendors and end users. Contributions to standards
are generated by companies, and they are first submitted to national technical coordination
groups, resulting in national standards. These national coordinating bodies may also pass
on contributions to regional organizations, or directly to ITU-T, resulting in regional
or world standards. ITU more recently started recommending and referencing standards
adopted by the other groups, instead of rewritingthem.
ITU-T is organized into 15 technical study groups. At present, more than 2500 recom-
mendations (standards) or some 55 000 pages are in force. They are nonbinding stan-
dards agreed by consensus in the technical study groups. Although, nonbinding, they are
generally complied with due to their high quality, and also because they guarantee the
interconnectivity of networks, and enable telecommunications services to be provided on
a worldwide scale.
ITU-T standards are published as recommendations, and they are organized into series.
Each series of recommendations is referred to by a letter of the alphabet. Some of the
well-known recommendations are the I, Q and X. Recommendations I are related to
integrated services digital networks. For instance, 1.321 describes the B-ISDN protocol
reference architecture, 1.370 deals with congestion management in frame relay, and 1.371
deals with congestion management in ATM networks. Recommendations Q are related

6 INTRODUCTION
to switching and signaling. For instance, Q.2931 describes the signaling procedures used
to establish a point-to-point ATM switched virtual connection over the private UNI, and
Q.2971 describes the signaling procedures used to establish a point-to-multipoint ATM
switched virtual connection over the private UNI. Recommendations X are related to data
networks and open system communication. For instance, X.700 describes the management
framework for the OSI basic reference model, and X.25 deals with the interface between
a DTE and a DCE terminal operating in a packet mode and connected to a public data
network by a dedicated circuit.
The International Organizationfor Standardization (ISO)
ISO is a worldwide federation of national standards bodies from some 130 countries, one
from each country. It is a nongovernmental organization established in 1947. Its mission
is to promote the development of standardization and related activities in the world, with
a view to facilitating the international exchange of goods and services, and to developing
cooperation in the spheres of intellectual, scientific, technological and economic activity.
It is interesting to note that the name ISO does not stand for the initials of the full title
of this organization, which would have been IOS! In fact, ISO is a word derived from the
Greek isos, which means 'equal'. From 'equal' to 'standard' was the line of thinking that
led to the choice of ISO. In addition, the name ISO is used around the world to denote
the organization, thus avoiding a plethora of acronyms resulting from the translation of
'International Organization for Standards' into the different national languages of the ISO
members, such as IOS in English, and OIN in French (from OrganizationInternational
de Normalization).
ISO's standards covers all technical fields. Well known examples of ISO standards
are: the ISO film speed code, the standardized format of telephone and banking cards,
ISO 9000 which provides a framework for quality management and quality assurance,
paper sizes, safety wire ropes, ISO metric screw threads, and the ISO international codes
for country names, currencies and languages. In telecommunications,the Open System
Interconnection (OSI) reference model (see Chapter 2) is a well known ISO standard.
ISO has co-operated with the International Electronical Commission (IEC) to develop
standards in computer networks. IEC emphasizes hardware, while ISO emphasizes soft-
ware. In 1987 the two groups formed the Joint Technical Committee 1 (JTC 1). This
committee developed documents that became ISO and IEC standards in the area of infor
mation technology.
The American National Standards Institute (ANSI)
ANSI is a nongovernmental organization formed in 1918 to act as a cross between a
standards setting body and a coordinating body for US organizations that develop stan-
dards. ANSI represents the US in international standards bodies such as ITU-T and ISO.
ANSI is not restricted to information technology. In 1960 ANSI formed X3, a committee
responsible for developing standards within the information processing area in the US. X3
is made up of 25 technical committees, of which X3S3 is the committee responsible for
data communications. The main telecommunications standards organization within ANSI
is the Tl secretariat, sponsored by the Exchange Carriers Standards Association. ANSI is
focused on standards above the physical layer. Hardware oriented standards are the work
of the Electronics Industries Association (ElA) in the US.

STANDARDS COMMITTEES 7
The Institute of Electrical and Electronics Engineering (IEEE)
IEEE is the largest technical professional society in the world, and it has been active in
developing standards in the area of electrical engineering and computing through its IEEE
Standards Association (IEEE-SA). This is an international organization with a complete
portfolio of standards. The IEEE-SA has two governing bodies: the Board of Governors,
and the Standards Board. The Board of Governors is responsible for the policy, financial
oversight, and strategic direction of the Association. The Standards Board has the charge
to implement and manage the standards process, such as approving projects.
One of the most well known IEEE standards bodies in the networking community is
the LAN/MAN Standards Committee, or otherwise known as the IEEE project 802. They
are responsible for several well known standards, such as CSMA/CD, token bus, token
ring, and the Logical Link Control (LLC) layer.
The Internet Engineering Task Force (IETF)
The IETF is part of a hierarchical structure that consists of the following four groups: the
Internet Society (ISOC) and its Board of Trustees, the Internet Architecture Board (IAB),
the Internet Engineering Steering Group (IESG), and the Internet Engineering Task Force
(IETF) itself.
The ISOC is a professional society concerned with the growth and evolution of the
Internet worldwide. The IAB is a technical advisory group of the ISOC, and its charter
is to provide oversight of the Internet and its protocols, and to resolve appeals regarding
the decisions of the IESG. The IESG is responsible for technical management of IETF
activities and the Internet standards process. It administers the standardization process
according to the rules and procedures which have been ratified by the ISOC Trustees.
The IETF is a large open international community of network designers, operators,
vendors and researchers concerned with the evolution of the Internet architecture and the
smooth operation of the Internet. It is divided into the following eight functional areas:
applications, Internet, IP: next generation, network management, operational requirements,
routing, security, transport, and user services. Each area has several working groups. A
working group is made up of a group of people who work under a charter in order to
achieve a certain goal. Most working groups have a finite lifetime, and a working group
is dissolved once it has achieved its goal. Each of the eight functional areas has one or
two area directors, who are members of IESG. Much of the work of IETF is handled via
mailing lists, which anyone can join.
The IETF standards are known as Request For Comments (RFC), and each of them is
associated with a different number. For instance, RFC 791 describes the Internet Protocol
(IP), and RFC 793 the Transmission Control Protocol (TCP). Originally, an RFC was
just what the name implies, that is, a request for comments. Early RFCs were messages
between the ARPANET architects about how to resolve certain procedures. Over the
years, however, RFCs became more formal, and they were cited as standards, even when
they were not. There are two subseries within the RFCs, namely, For Your Information
(FYI) RFCs and standard (STD) RFCs. The FYI RFC subseries was created to document
overviews and topics which are introductory in nature. The STD RFC subseries was
created to identify those RFCs which are in fact Internet standards.

8 INTRODUCTION
Another type of Internet document is the Internet-draft. These are work-in progress
documents of the IETF, submitted by any group or individual.These documents are valid
for six months, and they may be updated, replaced, or they may become obsolete.
Finally, we note that the ISOC has also chartered the Internet Assigned Numbers
Authority (IANA) as the central coordinator for the assignment of 'unique parameters'
on the Internet, including IP addresses.
The ATM Forum
During the late 1980s, many vendors felt that the ATM standardization process in ITU-T
was too slow. The ATM Forum was created in 1991 with the objective of accelerating
the use of ATM products and services in the private domain through a rapid development
of specifications. The ATM Forum is an international, nonprofit organization, and it has
generated very strong interest within the communications industry. Currently, it consists
of over 600 member companies, and it remains open to any organization that is interested
in accelerating the availability of ATM-based solutions.
The ATM Forum consists of the Technical Committee, three Market Awareness Commit-
tees for North America, Europe and Asia-Pacific, and the User Committee.
The ATM Forum Technical Committee works with other worldwide standards bodies
selecting appropriate standards, resolving differencesamong standards, and recommending
new standards when existing ones are absent or inappropriate. It was created as a single
worldwide committee in order to promote a single set of specifications for ATM prod-
ucts and services. It consists of several working groups, which investigate different
areas of ATM technology, such as the ATM architecture, routing and addressing, traffic
management, ATM/IP collaboration, voice and multimedia over ATM, control signaling,
frame-based ATM, network management, physical layer, security, wireless ATM. and
testing.
The ATM Market Awareness Committees provide marketing and educational services
designed to speed the understanding and acceptance of ATM technology. They coordinate
the development of educational presentation modules and technology papers, publish
the 53 Bytes, the ATM Forum's newsletter, and coordinate demonstrations of ATM at
trade shows.
The ATM Forum User Committee, formed in 1993, consists of organizations which
focus on planning, implementation, management or operational use of ATM-based net
works, and network applications. This committee interacts regularly with the Market
Awareness Committees and the Technical Committee to ensure that ATM technical spec-
ifications meet real-world end-user needs.
The Frame Relay Forum
The Frame Relay Forum was formed in 1991, and is an association of vendors, carriers,
users and consultants committed to the implementation of frame relay in accordance with
national and international standards.
The Forum's technical committees take existing standards, which may not be sufficient
for full interoperability, and create Implementation Agreements (IA). These lAs represent
an agreement by all members of the frame relay community as to the specific manner in
which standards will be applied. At the same time, the Forum's marketing committees

PROBLEMS 9
are chartered with worldwide market development through education as to the benefits if
frame relay.
PROBLEMS
1, Visit the web sites of ITU-T, the ATM Forum and IETF. Familiarize yourself with their orga-
nizational structure, and the type of standards that are available on these web sites.
2. Read some of the issues of 53 Bytes, the ATM Forum's newsletter, available on the ATM
Forum's web site.

Basic Concepts from Computer
Networks
In this chapter, we review some basic concepts from computer networks that we use in
this book. First, we discuss the various communication networking techniques and the
OSI reference model. Then, we present the data link layer of the OSI model, the High-
level Data Link Control (HDLC), the synchronous Time Division Multiplexing (TDM)
technique, and the Logical Link Control (LLC) layer. Finally, we examine the network
access protocol X.25, and conclude with the very popular and important Internet Protocol
version 4 (IPv4).
2.1 COMMUNICATION NETWORKING TECHNIQUES
Communication networking techniques can be classified into the following two broad cate-
gories: switched and broadcast communication networks. Examples of switched commu-
nication networks are circuit-switched networks, such as the public telephone system, and
packet-switched networks, such as computer networks based on TCP/IP. Examples of
broadcast communication networks are packet radio networks, satellite networks, and
multi-access local networks such as Ethernet. ATM networks belong to the packet-
switched networks.
Circuit switching and packet switching are two different technologies that evolved over
a long period of time. Circuit switching involves three phases: circuit establishment, data
transfer and circuit disconnect. These three phases take place when we make a phone
call. Circuit establishment takes place when we dial up a number. At that moment, the
public network attempts to establish a connection to the phone set that we dialed. This
involves finding a path to the called party, allocating a channel on each transmission link
on the path, and alerting the called party. The data transfer phase follows, during which
we converse with the person we called. Finally, the circuit disconnect phase takes place
when we hang up. At that moment, the network tears down the connection, and releases
the allocated channel on each link on the path. In circuit switching, channel capacity
is dedicated for the duration of the connection, even when no data is being sent. For
instance, when we make a phone call, the channel that is allocated on each transmission
link along the path from our phone to the one we called is not shared with any other
phone calls. Also, in circuit switching both stations must be available at the same time
in order to establish a connection. Circuit switching is a good solution for voice, since
It involves exchanging a relatively continuous flow of data. However, it is not a good
solution if the data is bursty. That is, the source emitting the data is active transmitting
1
2

12 BASIC CONCEPTS FROMCOMPUTERNETWORKS
for a period of time, and then it becomes silent for a period of time during which it is not
transmitting. This cycle of being active and then silent repeats until the source completes
its transmission. Such an intermittent type of transmission occurs in data transfers. In
such cases, the utilization of the circuit-switched connection is low.
Packet switching is appropriate for data exchange. Information is sent in packets, and
each packet has a header with the destination address. A packet is passed through the
network from node to node until it reaches its destination. Error and flow control proce-
dures can be built into the network to ensure a reliable service. In packet switching, two
different techniques can be used, virtual circuits and datagrams.
A virtual circuit imitates circuit switching, and it involves the same three phases:
call set-up, transfer of packets, and call termination. In call set-up, a logical connection is
established between the sender and the receiver before any packets are allowed to be sent.
This is a path through the nodes of the computer network which all packets will follow.
Unlike circuit switching, channel capacity on each transmission link is not dedicated to a
virtual circuit. Rather, the transmission link is shared by all the virtual circuits that pass
through it. Error control ensures that all packets are delivered correctly in sequence. Flow
control is used to ensure that the sender does not over-run the receiver's input buffer.
The X.25 network is a good example of a packet-switched network with virtual circuits.
Also, as we will see in Chapter 4, ATM networks are also packet-switched networks, and
they use virtual circuits.
In datagrams, no call set-up is required, and each packet is routed through the network
individually. Because of this, it is possible that two successive packets transmitted from
the same sender to the same receiver may follow different routes through the network.
Since each packet is routed through the network individually, a datagram service can
react to congestion easier. The datagram service provided by the early packet-switched
networks was in some cases more primitive than that provided by virtual circuits. For
instance, there was no error control, no flow control, and no guarantee of delivering
packets in sequence. The IP network, used in the Internet, is a packet-switched network
based on datagrams. However, due to the use of static routes in the IP routers, IP packets
follow the same path from a sender to a destination, and therefore they are delivered in
sequence. Also, unlike earlier packet-switched networks with datagram services, TCP/IP
provides both error and flow control.
An example of how two nodes communicate using circuit switching, virtual circuits,
and datagrams is given in Figure 2.1. In this example, node 1 communicates with node
4 through intermediate nodes 2 and 3. The passage of time is indicated on the vertical
lines, and there is one vertical line per node. In the circuit switching case, the time it
takes node 1 to transmit the call request packet and the message is indicated vertically
between the two arrows on the first line associated with node 1.The two diagonal parallel
lines between the vertical lines of the first and the second nodes show the propagation
delay of the call request packet between these two nodes. Similar notation is used for the
virtual circuit and datagrams cases. As we can see, the datagram scheme takes less time
to transmit the three packets than the virtual circuit scheme.
A broadcast network has a single communication channel that is shared by all the
stations. There are no switching nodes as in circuit or packet switching. Data transmitted
by one station is received by many, and often by all. An access control technique is
used to regulate the order in which stations transmit. The most widespread example of a
broadcast network is the Ethernet.

THE OPEN SYSTEM INTERCONNECTION (OSI) REFERENCEMODEL 13
Figure 2.1 A comparison between circuit-switching, virtual circuits and datagrams.
2.2 THE OPEN SYSTEM INTERCONNECTION (OSI) REFERENCE MODEL
In the early days of packet switching, the various communications software suites that
were available could not communicate with each other. To standardize the communications
protocols, and also facilitate their development, the International Organization for Stan-
dardization (ISO)proposed a model known as the Open Systems Interconnection (OSI)
Reference Model. The functionality of the software for packet switching was grouped
into seven layers, namely, the physical layer, the data link layer, the network layer, the
transport layer, the session layer, the presentation layer, and the application layer. These
layers are shown in Figure 2.2. Each layer provides service to the layer directly above it,
and receives service from the layer directly below it.
The physical layer is concerned with the transmission of raw bits over a communica-
tions channel. The data link's function is to transform the raw transmission link provided
by the physical layer into a reliable communications link. This was deemed necessary
since early transmission links were inherently unreliable. Modern fiber-based commu-
nications links are highly reliable, and as will be seen later on in this book, there is
no need for all the data link functionality. The network layer is concerned with routing
packets from source to destination, congestion control, and internetworking. The transport
protocol is concerned with the end-to-end packet transfer, that is, between an application
in the source computer and an application in the destination computer. Some of its main
functions are establishment and deletion of connections, reliable transfer of packets, and
flow control. The session layer allows users in different computers to set up sessions

14 BASIC CONCEPTS FROM COMPUTER NETWORKS
Figure 2.2 The OSI reference model.
between themselves. One of the services of the session layer is to manage dialogue
control. The presentation layer is concerned with the syntax and semantics of the infor-
mation transmitted. In general, two heterogeneous computers may not have the same way
of representing data types internally. The presentation layer facilitates the communication
between two such computers, by converting the representation used inside a computer to a
network standard representation and back. Finally, the application layer contains protocols
that are commonly used, such as file transfer, electronic mail and remote job entry.
2.3 DATA LINK LAYER
This protocol layer was designed to provide a reliable point-to-point connection over an
unreliable link. The main functions of the data link layer are: window flow control, error
control, frame synchronization, sequencing, addressing, and link management. At this
layer, a packet is referred to as a frame. Below, we examine the window-flow control
mechanism, error detection schemes, and the error control mechanism.
Window-flow control
This is a technique for ensuring that a transmitting station does not over-run the receiving
station's buffer. The simplest scheme is stop-and-wait. The sender transmits a single frame
and then waits until the receiver gets the frame and sends an acknowledgment (ACK).
When the sender receives the ACK, it transmits a new frame. This scheme is shown in
Figure 2.3. The link's utilization U depends on the propagation delay, tprop, and on the
time to transmit a frame, tframe.
Let
Then,

DATA LINK LAYER 15
Figure 2.3 The stop-and-wait scheme.
If a <<< 1, that is the propagation delay is significantly less than the time to transmit a
frame, then the link's utilization U is large. If a>>>1, that is the propagation delay is
significantly greater than the time to transmit a frame, then U is small. As an example,
let us consider a satellite link transmitting at 56 Kbps, and let us assume 4000-bit frames
and a propagation delay of 270ms. Then, the time to transmit a frame is 71ms, a =
270/71 =3. 8, and U = 0.116.
In the stop-and-wait protocol, only one frame is outstanding (i.e. unacknowledged)
at a time. A more efficient protocol is the sliding window-flow control protocol, where
many frames can be outstanding at a time. The maximum number of frames, W, that a
station is allowed to send to another station without acknowledgment is referred to as
the maximum window. To keep track of which frames have been acknowledged, each
frame is numbered sequentially, and the numbers are reusable. An example of the sliding
window-flow control scheme is shown in Figure 2.4. The maximum window size W is
fixed to 8. In Figure 2.4(a), station A transmits four frames with sequence numbers 1, 2, 3
and 4, and its window is reduced to four, consisting of the sequence numbers {5, 6, 7, 8}.
In Figure 2.4(b), station A sends two more frames with sequence numbers 5 and 6, and
its window is down to two, consisting of the numbers {7, 8}. In Figure 2.4(c), station A
receives an ACK from station B for the frames with sequence numbers 1, 2 and 3, and
its window opens up to five frames consisting of the sequence numbers {7, 8, 1, 2, 3}.
The efficiency of this protocol depends upon the maximum window size and the round-
trip delay. Let tframe
= 1. Then,
The time to transmit the first frame and receive an acknowledgment is equal to tframe +
2tprop = 1+ 2a. If W > 1+ 2a, then the acknowledgment arrives at the sender before the

16 BASIC CONCEPTSFROMCOMPUTER NETWORKS
Figure 2.4 An example of the sliding window-flow control scheme.
window has been exhausted, and we have that U = 1. If W < 1+ 2a, then theacknowl-
edgment arrives after the window has been exhausted, and we have
Error detection
The simplest error detection scheme is the parity check. In this scheme, a parity bit is
appended to the end of each frame. A more complex error detection scheme based on the
parity check is the longitudinal redundancy check. The data is organized into a matrix, as
shown in Figure 2.5. There are eight columns, and as many rows as the number of bytes.
Each matrix element contains one bit. An even parity check is applied to each row and
each column. We observe that the parity bit applied to the last column, which contains
the parity bits of all the rows, is the same as that applied to the last row which contains
the parity bits of all the columns!
The Cyclic Redundant Check (CRC) is a commonly used error detection scheme, and is
used extensively in ATM networks. The CRC scheme utilizes a predetermined bit pattern
P, which is known to both the sender and the receiver. Let n + 1 be the length of this bit
pattern. Now, let us assume that we have a k-bit message M to be transmitted. The sender
shifts M to the left by n bits to obtain the quantity 2n
M, and then divides 2n
M by P.
The remainder of that division is an n-bit sequence, known as the Frame CheckSequence
(FCS). The FCS is added to 2n
M and the entire (k + n)-bit message is transmitted to the
Figure 2.5 The longitudinal redundancy check.

DATA LINK LAYER 17
receiver. The receiver divides the message by the same bit pattern P. The message has
been received correctly if the remainder of that division is zero. All single bit errors, and
some combinations of erroneous bits, can be detected and corrected.
As an example let M = 1010001101 and P = 110101. Then, the FCS will be five bits
long and it is calculated as follows. M is first shifted to the left by five positions, that
is 25
M = 101000110100000. Then, 25
M is divided by P, resulting in an FCS equal to
01110. Finally, the transmitted message is 101000110101110. If this message is correctly
received, when divided by P = 110101, it should give a zero remainder.
It is customary to express the bit pattern P in polynomial form. This is done as follows.
Each bit is represented by a term xn
, where n is the location of the bit in the pattern,
counting from the right-hand side towards the left-hand side. That is, the rightmost bit
corresponds to the term x0, the second rightmost bit corresponds to the term x1
and
so on. The value of the bit is the coefficient of its corresponding polynomial term. For
instance, the pattern 110101 used above is expressed as x5
+ x4
+ x2
+ 1.
The checksum is another error detection technique that is used in the TCP/IP suite
of protocols. The data to be sent is treated as a sequence of binary integers of 16 bits
each, and the sum of these 16-bit integers is computed. The data could be of any type
or a mixture of types. It is simply treated as a sequence of integers for the purpose of
computing their sum. The 16-bit half-words are added up using 1's compliment arithmetic.
The 1's compliment of the final result is then computed, which is known as the checksum.
32-bit integers can also be used. The checksum is used in TCP to protect the entire packet,
i.e. it is calculated using the header and the payload of the TCP packet. It also used in
IP to protect the IP header only. Computing the checksum in TCP is a time-consuming
operation, and a considerable speed up can be achieved if it is done in hardware.
Error control
Error control refers to the mechanism used to detect and correct errors that have occurred
in the transmission of frames. This mechanism is known as the Automatic Repeat Request
(ARQ), and it uses error detection, the window-flow control mechanism, positive and
negative acknowledgments, and timers. Errors in the transmission of frames occur because
a frame is lost or because it is damaged, that is, one or more of its bits have been flipped.
Damaged frames are detected by the ARQ mechanism using CRC, and lost frames are
detected by observing out-of-sequence frames. Recovery of a lost or damaged frame is
done by requesting the sender to re-transmit the frame. Three different versions of the
ARQ have been standardized, namely stop-and-wait ARQ, go-back-n ARQ and selective-
reject ARQ. The stop-and-wait ARQ is based on the stop-and-wait window-flow control
scheme, whereas the go-back-n ARQ and the selective-reject ARQ are based on the sliding
window-flow control scheme.
In the go-back-n scheme, the sender sends a series of frames using the sliding window-
flow control technique. Let us assume that station A is transmitting to station B. If
B receives a frame correctly, then it sends an ACK with the next frame number that it
expects to receive. An ACK may be for several successive frames that have been correctly
received. If B receives a damaged frame, say frame i, and it has previously received
correctly frame i —1, then B sends a negative acknowledgment (NAK), indicating that
frame i is in error. When A receives the NAK, it retransmits frame i plus all other frames
after i that it has already transmitted. An example of this scheme is shown in Figure 2.6.

18 BASIC CONCEPTS FROM COMPUTERNETWORKS
Figure 2.6 The go-back-n scheme.
Now, let us consider the case where frame i is lost. If B correctly receives frame
i + 1 later on, then it will realize that frame i + 1 is out-of-sequence, and it will deduce
that frame i is lost. B will then send a NAK, indicating that the ith frame has to be
retransmitted. A retransmits frame i plus all other frames after i that it has already
transmitted. If frame i is lost and no other frames arrive, then B cannot detect the lost
frame. However, for each transmitted frame, A sets a timer. If the timer expires before A
receives an ACK or a NAK, A retransmits the frame. In the above case, the lost frame's
timer will expire and A will re-transmit it.
In the selective-reject ARQ scheme, only the frame that is in error is retransmitted.
All subsequent frames that arrive at B are buffered, until the erroneous frame is received
again. This is a more efficient procedure, but it is more complex to implement. The
selective-reject scheme is used in TCP.An example of the selective-reject ARQ scheme
is shown in Figure2.7.
2.4 THE HIGH DATA LINK CONTROL (HDLC) PROTOCOL
This protocol has been widely used, and it has been the basis for many other important
data link protocols. It was derived from IBM's data link protocol Synchronous Data Link
Control (SDLC). Later on it was modified and standardized by ISO as the High DataLink
Control (HDLC) protocol. HDLC was designed to satisfy different types of stations, link
configurations and transfer modes. The following three types of stations were defined:
primary, secondary and combined. A primary station is responsible for controlling the
operation of the link, a secondary station operates under the control of a primary station,
Figure 2.7 The selective-reject scheme.

THE HIGH DATA LINK CONTROL (HDLC) PROTOCOL 19
and a combined station has the features of both the primary and the secondary station.
Also, the following types of link configurations were defined: unbalanced and balanced.
An unbalanced configuration consists of one primary and one or more secondary stations,
and it supports both full-duplex and half-duplex transmission. A balanced configuration
consists of two combined stations, and it supports both full-duplex and half-duplex trans-
mission. Based on these station types and configurations, the following three data transfer
modes were defined: Normal Response time Mode (NRM), Asynchronous Balanced Mode
(ABM), and Asynchronous Response Mode (ARM). NRM is used with an unbalanced
configuration. The primary station initiates data transfers to the secondary stations, and
a secondary station may only transmit data in response to a command from the primary.
NRM is used in multi-drop lines connecting terminals to a host. ABM is used with a
balanced configuration, and it is the most widely used transfer mode for a full-duplex
point-to-point link. Either combined station may initiate a transmission without receiving
the permission from the other combined station. Finally, ARM is based on an unbalanced
configuration, and it is rarely used.
HDLC is a bit-oriented protocol, and it uses the frame structure shown in Figure 2.8.
A single format is used for all data and control exchanges. The frame is delimited by a
flag which contains the unique pattern 01111110. If frames are transmitted back-to-back,
a single flag may be used to indicate the end of one frame and the beginning of the
next one. Obviously, the pattern 01111110 can be easily encountered within a frame, in
which case it will be interpreted as the end of the frame. To avoid this from happening,
a technique known as bit stuffing is used. The sender always inserts an extra 0 after the
occurrence of five consecutive 1's. The receiver monitors the bit stream looking for five
consecutive 1's. When this pattern appears, the receiver examines the sixth bit. If it is a
0, it is deleted from the bit stream. If it is a 1 and the seventh bit is a 0, the receiver
interprets the bit pattern as a delimiting flag. If the sixth bit is a 1 and the seventh bit is
also a 1, then it is an error.
The second field in the HDLC frame is the address field. This is an 8-bit field used
in multi-drop lines, and it is used to identify the secondary station to which the frame is
transmitted. It is not necessary in a point-to-point link.
The third field in the HDLC frame is the control field. It is an 8-bit field, extendible
to a 16-bit field, and its structure is shown in Figure 2.9. It is used to identify the
following three types of frame: information frame (I-frame), supervisory frame (S-frame),
and unnumbered frame (U-frame). An I-frame is used to carry data and ARQ control
information, an S-frame is used to carry only ARQ control information, and a U-frame
is used to provide supplemental link control functions. If the first bit of the control field
is 0, then the frame is an I-frame. Otherwise, depending on the value of the second
bit, it may be an S-frame or a U-frame. The meaning of the remaining sub-fields is as
follows:
Figure 2.8 The HDLC frame.

Exploring the Variety of Random
Documents with Different Content

His body was doubled so that he could not use his hands to tear the
bag or strike out.
In two minutes he had relinquished all hope.
He began to wish that he had never heard of the Mahdi, or the
Mameluke.
But regrets were useless.
He knew he had to die.
Had it been on the battlefield, pitted against a foe, he would have
been proud to die—because he knew no disgrace would be attached
to it.
But to die in a sack, like a mangy dog or vicious cat, was so hurtful to
his self-respect and so humiliating that he cried with vexation.
The water got to his lungs. His stomach was full of it. His brain grew
dizzy.
The singing in his ears had become like the roaring of the waters of
a great cataract.
Mercifully unconsciousness came, and had not the conspirators
been discussing their schemes of rioting and rebellion at night by the
banks of the Nile, Madcap Max would never have been the hero of
this story.
Shula rubbed Max briskly.
He straightened out the madcap’s body and laid it face downward.
The conspirators began kneading the poor fellow’s back—sitting on
it, treading it, kneeling on it, and using every means of which they
knew to restore life.
“Get out of that and meet a fellow face to face.”
The words startled the conspirators.
They were uttered by Max, who, black and blue with the treatment
he had been subjected to, had revived with great suddenness.

He did not realize where he was, but he knew he was being hurt,
hence his calling out.
He jumped to his feet.
“Shula!” he exclaimed.
“Max!”
“Yes. How did you find me? Was I drowned? Where am I?”
“You are not drowned; you are by the Nile’s water, and the less you
say the longer you will be likely to live. Come—let us get home. Can
you walk?”
“Of course I can.”
Max started forward, but before his legs had moved a dozen times
he fell on his face.
The conspirators lifted him up, and as no conveyances were to be
found in Kordofan at that hour of the night, they had to carry him to
Shula’s residence.
Before morning’s dawn he had told his adventures and laughed at
the escapade.
“If ever the Mahdi rules in Kordofan I am going to see Lalla,” he said.
“I want to know more about her.”
“Not even the prophet could give you the right to enter any man’s
harem,” said Shula.
“Then your Mahdi must be a queer sort of fellow.”
Max was unable to talk longer, for he was naturally weak from his
struggles in the Nile.
Twenty-four hours elapsed before he was able to feel that he was
the strong athlete again.
When he awoke on the morning of the third day he heard cries which
roused him:
“Allah il Allah!”
“Long live the Mahdi!”

“Down with the foreigner!”
“The Mahdi has come!”
Max looked at Shula, but the merchant did not speak.
His face was white as that of a corpse. He knew that he had staked
all his property and his life on the riot which was then in progress.
“Is it true? Has the Mahdi come?”
“No, Max, but the people are expecting him.”
A heavy fusillade was heard on the streets, the windows were
shaken, and some panes of glass broken.
“What does it mean?”
“They are fighting,” answered Shula.

CHAPTER XXX.
THE MAHDI’S JUSTICE.
“Fighting, and you here? Why are not you at the head of the Mahdi’s
friends?”
“I—stayed—with you.”
“Come! where is my sword?”
“It is here; but don’t go out. You will be killed—the soldiers wouldn’t
join the Mahdi, and they are shooting the people down.”
“Give me my Winchester and my sword.”
“It is madness.”
“Well, I am the madcap,” laughed Max; “but if I wasn’t I’d scorn to be
a coward.”
“A coward?”
“Yes, I said so, and I repeat—a coward.”
“Why do you call me that? I have fought in the army of Egypt.”
“Perhaps so. But did you not stir up this riot and are now afraid——”
“I am not afraid; but is it policy to risk so much?”
“Risk all—if by that means you save your honor.”
“But the people have no chance against the soldiers.”
“All the more reason why you should not desert them.”
“See what it means to me—loss of property, perhaps life.”
“Do as you like, most excellent Shula, but I am going to fight.”
“It is madness!”
“Give me my rifle and my sword.”
Max seized the weapons and rushed into the street.

He saw the rioting, and felt that Shula was right—the people had but
scant chance.
That made Max all the more determined.
He waved his sword above his head and rushed into the thickest of
the fight.
“Long live the Mahdi!”
At the sight of the paleface the soldiers fell back.
“I am an American,” shouted Max, “but I am with you. The Mahdi is a
native of your country, he is no foreigner. Strike for him, and let your
cry be Egypt for the Egyptian, the Soudan for the Soudanese!”
The people lost their fear.
Like demons they sprang on the soldiers, but the soldiers did not
return the fire.
Instead, they reversed their guns and retired.
The Egyptian officer was enraged.
“I’ll shoot the first man who deserts!” he shouted.
A number of the soldiers again shouldered arms, but the majority
kept them reversed.
Max saw the advantage he had gained.
He caught the bridle of a horse whose rider had fallen in the mêlée.
Vaulting into the saddle, he looked proud and defiant as he sat there,
like a veritable centaur.
“Soldiers, you believe in Mahomet! Hark ye! I have fought with the
great Mahdi. I have seen the thousands of Fashoda beaten back
when he waved his wand. He has no need of sword or scimiter; he
fights with his eyes, and when he waves his hand, armies fall back.”
The enthusiasm was great.
Max had won over most of the soldiers, and the others were
undecided.

The officer was furious.
“Ready!” he shouted, but very few of his men obeyed the call.
“Load! Aim! Fire!”
Half a dozen rifle shots were fired, but Max saw to his great joy that
the aim was too high to do any damage.
“Men! soldiers of the crescent!” he called out, “our fight is not against
you. The Mahdi is of your faith. Nay, more, he will restore the great
Mameluke kingdom. Every soldier of his will be greater than a pasha,
for the Mahdi is the last of the Mamelukes.”
The speech was listened to by soldiers and people, who wondered
who this young paleface could be.
The result was electrical.
Every rifle was reversed.
The officer was left alone to return to the fort—a commander without
soldiers.
At the time when Max so eloquently proclaimed the Mahdi,
Mohammed Achmet was close to the gates of the city. He heard the
cheering and the firing.
His face paled visibly, for he disliked bloodshed.
Half an hour later, riding between the Persian Sherif el Habib and the
Arab Mohammed, the Mahdi rode into the main street of Kordofan.
“The Mahdi!”
“The Mahdi has come!”
The cheers rose on the air.
Songs were sung—the soldiers fraternized with the people.
Everywhere the enthusiasm was intense.
Even the garrison joined in the cheering, and the officer handed his
sword to the Mahdi.

“I cannot fight without men,” he said, “so take my sword and use it
for truth and our faith.”
The Mahdi took the weapon, and immediately handed it back,
saying:
“General, you are a brave man. Take the sword, for you will use it as
only a brave man can.”
The fires of joy were lighted.
Houses were thrown open, and everywhere the Mahdi was
welcomed.
Mahmoud Achmet, when he saw that the Mahdi was triumphant,
came to offer the hospitality of his house to the conqueror.
Max recognized him, and after the man had said all he intended,
came forward.
“You threw a young man into the Nile. You enveloped him in a sack,
and drowned him.”
“It is he! I know it! The Mahdi is the Mahdi. He has raised this man
from the dead. All my wealth is his,” exclaimed Mahmoud.
Max saw the mistake the man had made. He, however, did not
contradict him, but allowed him to think that the power of the Mahdi
had indeed raised him from the dead.
He spoke privately to the Mahdi.
“Let him give me Lalla,” said Max.
“You spoke of your wealth,” said the Mahdi; “give this man the girl
called Lalla.”
Mahmoud fell to the ground.
He tore his hair and pulled out his beard.
“Woe is me, I cannot!”
“She is dead?” queried the Mahdi.
“Indeed it is true. Inshallah!”

Mahmoud then admitted that he was jealous of Max, and after
throwing him into the river, Lalla had refused to be comforted, had
called him a murderer, and refused to allow him to approach her.
Then it was that in his anger he ordered her to be drowned.
Max told of the brutal way in which Mahmoud acted.
The Mahdi called the pashas and beys together, and in the presence
of a great concourse of citizens, said:
“One of your number, Mahmoud Achmet, has at times made away
with such of his wives that displeased him. Now, therefore, to prove
to you how abhorrent such a thing is, it is my order that Mahmoud
Achmet be taken from here in the sack which he has provided for
others, and that he be thrown into the Nile.”
“Mercy!” cried the wealthy man—“mercy! I will give you wealth.”
“I do not want it.”
“All I have shall be yours!”
“It is mine already.”
One of the eunuchs connected with Mahmoud’s harem testified how
the wives were constantly beaten with whips.
“The same measure shall be meted out to Mahmoud,” said the
Mahdi; “it is fate.”
The man pleaded for his life, but the Mahdi was inexorable.
Mahmoud suffered the scourging from the hands of his own eunuch,
and was drowned in the Nile.
“It is fate! It is justice!” exclaimed the people, who were more than
ever enthused with the prophet and his cause.

CHAPTER XXXI.
VICTORY ALL ALONG THE LINE.
Early on the following morning a man, riding at hot haste, asked for
the Mahdi.
He bore a letter to the prophet, and another to Sherif el Habib.
When the dispatch was opened the Mahdi read:
“To the illustrious Mahomet Ahmed, the Prophet, Imaum
and Mahdi:
“Greeting: Senaar resisted for several hours, but the flag
of the Mahdi floats over its fortress. The day is ours.
“Ibrahim.”
Sherif el Habib handed his document to the Mahdi.
“Dear uncle, we have fought and won,” ran the letter. “I
was wounded in the right foot and lost two toes, but that
was better than my life. The people were all with us, but
the soldiers fought bravely. It was a tough battle. The
commander gave me his sword, which I will send to the
Mahdi when I hear from him. How is Girzilla? Give her my
love. Is Max the Madcap alive? Of course he is. Tell him
not to play any pranks in Kordofan.
“Your loving nephew,
“Ibrahim.”
When the Mahdi had read the letters aloud to his staff, he called Max
to him.
“It was your plan which we adopted,” he said, “and we are victorious.
You are Max Pasha; and your nephew”—turning to Sherif—“is also
pasha, and is made governor of Senaar, while Max, here, shall be
governor of Kordofan.”
The people cheered the young governor.

Turning to the Mahdi, Max said:
“I thank you for the honor, but I am about to decline it.”
“You must not.”
“I am about to decline it after to-morrow. I want to be governor and
pasha for one day, because I am going back to America, and if I ever
go on the lecture platform the people will sooner pay a dollar to hear
a real live pasha, than a quarter if the speaker is only Madcap Max.”
The Mahdi laughed.
“Still thinking of the dollars?” he said.
“Yes,” answered Max; “and whenever you get tired of being the
Mahdi come over to New York and I will trot you round, and—oh, my!
won’t the dollars just flow into our pockets.”
But before the Mahdi could reply another dispatch was placed in his
hands.
It was from a trusty agent in the North.
“Giegler Pasha has placed the army of Khartoum under the
command of Yussuf Pasha Hassan,” it read, “and is marching with
five thousand men against you. Hicks Pasha, an Englishman, with
three thousand men, is marching from the northeast. You are to be
cut in two by these armies.”
“No! by the prophet—no!” exclaimed the Mahdi. “We will attack both
and exterminate them.”
The bugles called the army together and the march was ordered.
With a speed accelerated by the most fanatical enthusiasm, the
followers of the Mahdi started to meet Yussuf Pasha Hassan.
The soldiers of Khartoum were well disciplined veterans, but they
lacked enthusiasm.
The Mahdi—still without weapon—rode at the head of his people
and gave the words of command.

Like a cyclone tearing everything before it on a Western prairie, the
army of the Mahdi swept on the veterans commanded by Yussuf.
The Egyptians made a stubborn resistance at first, but the Mahdists
were more like fiends.
They seized the soldiers by their hair and deliberately cut their
throats.
It was a horrible carnage.
The Mahdi never struck a blow, never made any effort to defend
himself, but was ever in the thickest of the fight.
His brow shone as though it were gold.
His presence was remarkable.
Max fought with desperate valor.
At times he stood up in the stirrups to give himself more power in
striking a blow.
“The Mahdi forever!” he shouted, with every savage blow.
Yussuf saw the young fellow and knew that, next to the Mahdi, Max
was the most powerful leader.
Yussuf would not touch the Mahdi.
He was a trifle superstitious.
If Mohammed was the Mahdi, steel weapons could not kill him, and
Yussuf would not risk an encounter; so he rode through the fighting
demons until he reached the side of Max.
“The Mahdi forever!” shouted Max, as he suddenly wheeled round
and aimed a blow at Yussuf’s head.
The veteran officer parried the blow and made a lunge at Max.
But the American’s sword swung round with cyclonic speed, and
Yussuf’s sword merely struck the air.
As the heavy scimiters clashed together sparks of fire flew out, and
seemed to keep fiery time to the music of the steel.

Yussuf got angry.
“Do you also bear a charmed life?” he sneeringly asked, during a
pause in the duel.
“I am an American,” answered Max, “and fight for liberty.”
Again the fight was resumed.
Great heaps of dead were to be found in every direction.
The horses ridden by Yussuf and Max often had to kick and trample
down the dead and dying.
It was a fearful sight.
Yussuf fought bravely.
His left arm had been broken by Max, just below the shoulder, but he
would not give in.
“Surrender!”
“Never!”
“Then die!”
“I will, but you will go first.”
Max was of a different opinion, and he kept swinging round his
heavy scimiter with the strength of a giant.
Once, when Yussuf parried a blow, the weapon struck the horse’s
neck, almost severing the head from the body.
Yussuf was now at a disadvantage.
Max leaped from the saddle and stood by the Egyptian’s side.
“We are equal,” he said.
But it was scarcely the truth, for Yussuf had only one arm to fight
with.
The Egyptian slipped in a pool of blood, and as he did so a sword
still grasped by a dead man pierced his side.
The brave man could stand no more.

“I surrender!” he gasped, but it was not a surrender to Max, but to
the Great Creator, for as the man uttered the words the breath left
his body.
Out of four thousand seven hundred men—hale, hearty veterans—
who had marched under the crescent of Egypt that morning, only two
hundred and one survived at night.
The Mahdists did not lose more than four hundred men all told.
They did not stop to care for the wounded or bury the dead.
Another blow had to be struck, and this time at Hicks Pasha.
It was a two days march to Tokar.
At that place Hicks, with three thousand seven hundred and forty-six
men, met the advance guard of the Mahdists, led by Sherif el Habib
and Max.
The fighting was desperate, but seemed to be as favorable to the
Egyptians as the Mahdists, until the Mahdi himself arrived.
There was a charm and magnetism about the man which made him
irresistible.
His presence was equal to a thousand men.
In less than an hour the unfortunate Hicks was dead, and two
thousand three hundred and seventy-three of his men lay stiffening
under the tropical sun.
The defeat was a thorough one.
The Mahdi was now master of all the Soudan except Khartoum and
Equatoria, over which Emin Bey presided.
The people flocked to the Mahdi’s tent.
Dervishes proclaimed him to be the promised Imaum. In the
mosques his name was mentioned with that of the prophet, and the
people prostrated themselves when reference was made to him.

CHAPTER XXXII.
“ALL’S WELL THAT ENDS WELL.”
A week of peace after the storm of war was delightful.
The army of the Mahdists was large enough to crush any force which
could be sent against it.
The officers took things easy.
Mohammed had brought his harem to the Mahdi’s headquarters, and
Ibrahim had received a furlough or leave of absence for two months.
This gave him plenty of time to be with Girzilla.
One day Girzilla sought out Max and whispered:
“I have found him.”
“Whom do you refer to?”
“The last of the Mamelukes.”
“And he is——”
“The Mahdi.”
“Are you sure, Girzilla?”
“Yes; by secret signs I discovered him, and he will restore the glories
of his race and bring the whole world to believe in Mahomet.”
Max went to the Mahdi and told him of his mission.
The tears came into the warrior prophet’s eyes as he heard Max tell
his story; how he had lost his father in the caves of the bandits, and
had been rescued by Girzilla.
When Max narrated how he had become enthused over the story of
the great Mameluke who escaped from Mohammed Ali, the Mahdi
embraced him.

“For my ancestors’ sake, you are doubly dear to me. Stay with me,
my son, and share in my triumph.”
“No—the work is done. I shall go back to my own land, and shall do
as other Americans have done before me—write a book, or tell on
the platform the story of the Mahdi, and the Mameluke.”
Max wanted to start at once, but Ibrahim pleaded with him to stay
until after his wedding with Girzilla.
This Max consented to do, and three weeks later a most impressive
wedding took place in the vestibule of a mosque at Kordofan.
The couple were united and blessed by the Mahdi.
The Imaum made some pertinent remarks, which were worthy of the
great prophet himself.
To Ibrahim, after praising his courage, he said:
“You have taken to yourself a wife. The Koran permits you to take
three others; but take my advice—cleave to the one. It is better, and
a new dispensation will so order. Treat Girzilla, not as others of our
race have been treated, but let her be your equal; for it is now written
that if you be faithful to her on earth the gates of Paradise will open
for you both, and she shall be your bride through all eternity.”
After spending the customary seven days in prayer and religious
observances, Ibrahim obtained permission to take his dusky bride on
a trip up the Nile in company with Max.
The cataracts were passed, and Cairo reached.
Girzilla pleaded so earnestly to continue the journey that her loving
husband accompanied her to Suez, where they bade farewell to
Madcap Max as the Peninsular and Oriental steamer steamed out of
the port.
Max had not noticed that it was the very vessel he had made the
journey on three years before.
He made himself known to the captain, and the tedium of the journey
was broken by the story of adventure told by the madcap.

When Max reached New York he found himself the head of the firm,
and the cares of business life caused him to relinquish the thought of
“coining dollars” on the lecture platform; but he made a solemn
promise to the author that some day he would tell him the story of his
life.
Two years passed, and the author asked the well-known and highly
respected merchant to tell the story.
“To-morrow come to us, be our guest for a week, and you shall know
all.”
“But——”
“My wife will welcome you as an old friend.”
Max had married a fairer woman than Girzilla, but many a time he
declared that no more true one ever lived than the Arab maiden.
When the author reached the Gordon uptown mansion on the
following day he was surprised to find so many evidences of the
Orient everywhere; but when, an hour later, Max took the author by
the hand and led him into a large parlor, he was still more surprised,
for there stood, waiting to receive him, Ibrahim and Girzilla.
Sherif el Habib was dead. His nephew had sold the shawl
manufactory, and found himself extremely wealthy.
He at once determined to make the “grand tour” of the world, and so
infatuated was he with the remembrance of Max, that nothing would
satisfy him but to commence the journey proper from New York.
That was how this story came to be written.
Max narrated it, but Ibrahim and Girzilla insisted on a more lavish
praise of the madcap than he would acknowledge he deserved.
Never was there a happier couple than the Persian and his lovely
bride, who does not look so dark and dusky in the modern American
clothing as she did on the deserts of Africa.

Ibrahim accepted the advice of the Mahdi, and declares that Girzilla
occupies every bit of his heart, and he could not take three more
wives, even if his religion ordered it.
Our story is told. All has ended happily for our madcap and his
friend, and although his heart turns sick sometimes as he thinks of
the carnage he witnessed, yet he says he shall always look back
with pride to the intimacy he had with Mohammed Ahmed, the Mahdi
and the Mameluke, the result of his trip “In the Volcano’s Mouth.”
THE END.

TALES OF VICTORIES
Gained in the Pre-Revolutionary wars by lads of pluck
and intelligence. Every true boy will be fascinated with
these stories of the exciting adventures of boys who
gladly gave their lives to freedom’s cause.
BOYS OF LIBERTY LIBRARY
3.—The Young Ambassador. By John De Morgan
7.—The Young Guardsman. By John De Morgan
11.—Fighting Hal. By John De Morgan
15.—By Order of the Colonel. By Lieut. Lounsberry
19.—A Call to Duty. By Lieut. Lounsberry
23.—The Young Patriot. By Lieut. Lounsberry
26.—The Trader’s Captive. By Lieut. Lounsberry
Only Ten Cents Per Copy
At All Newsdealers
If ordered by mail, add four cents to cover
postage.
STREET & SMITH, Publishers
NEW YORK

BOUND TO WIN
LIBRARY
A weekly publication devoted to high-class literature
for boys. Sept 14, 1905.
NO. 134
Charles Garvice’s New Stories
If you are a novel reader, you certainly must be waiting for the
appearance of a new novel from the pen of Charles Garvice. We
are glad to inform you that you will find it in
SMITH’S MAGAZINE
In the future, all of Charles Garvice’s new stories will appear in
this magazine, as he is under contract to write for it exclusively.
“DIANA’S DESTINY” is the title of a bright, original story, of
absorbing interest. It began in the April number and is still being
published.
If you are one of the vast army who have depended upon
cheap, occasional issues of early non-copyrights, of which there
are now no more by this author, you will find this new tale
distinctly refreshing.
In addition to a long installment of the Garvice story, there are
other features which make SMITH’S MAGAZINE one of the best
and most pleasing of all ten-cent publications.
PROFUSELY ILLUSTRATED BY THE BEST
ARTISTS
PRICE, TEN CENTS Sold by all Newsdealers
THE SMITH PUBLISHING HOUSE,
156 FIFTH AVENUE
NEW YORK

Transcriber’s Notes:
Punctuation has been made consistent.
Variations in spelling and hyphenation were retained as they appear in the original
publication, except that obvious typographical errors have been corrected.
The following change was made:
p. 211: Korfodan changed to Kordofan (street of Kordofan.)

Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!
ebookfinal.com

An Introduction to ATM Networks 1st Edition Harry G. Perros

More Related Content

Similar to An Introduction to ATM Networks 1st Edition Harry G. Perros (20)

Recently uploaded (20)

An Introduction to ATM Networks 1st Edition Harry G. Perros