An Introduction To Atm Networks 1st Edition Harry G Perros
- 1. An Introduction To Atm Networks 1st Edition Harry G Perros download https://ebookbell.com/product/an-introduction-to-atm- networks-1st-edition-harry-g-perros-1814632 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. An Introduction To Early Buddhist Soteriology Freedom Of Mind And Freedom By Wisdom G A Somaratne https://ebookbell.com/product/an-introduction-to-early-buddhist- soteriology-freedom-of-mind-and-freedom-by-wisdom-g-a- somaratne-44887694 An Introduction To Six Sigma And Process Improvement 2nd Edition James R Evans https://ebookbell.com/product/an-introduction-to-six-sigma-and- process-improvement-2nd-edition-james-r-evans-44913594 An Introduction To Conservation Biology 3rd Edition Richard B Primack https://ebookbell.com/product/an-introduction-to-conservation- biology-3rd-edition-richard-b-primack-44975620 An Introduction To Bioanalysis Of Biopharmaceuticals Seema Kumar https://ebookbell.com/product/an-introduction-to-bioanalysis-of- biopharmaceuticals-seema-kumar-44992290
- 3. An Introduction To Knot Theory Graduate Texts In Mathematics 175 Softcover Reprint Of The Original 1st Ed 1997 Wbraymond Lickorish https://ebookbell.com/product/an-introduction-to-knot-theory-graduate- texts-in-mathematics-175-softcover-reprint-of-the-original-1st- ed-1997-wbraymond-lickorish-44992914 An Introduction To The Theory Of Number 5th Edition Ivan Niven https://ebookbell.com/product/an-introduction-to-the-theory-of- number-5th-edition-ivan-niven-45136082 An Introduction To New Media And Cybercultures Pramod K Nayar https://ebookbell.com/product/an-introduction-to-new-media-and- cybercultures-pramod-k-nayar-45766740 An Introduction To The Mechanics Of Incompressible Fluids Michel O Deville https://ebookbell.com/product/an-introduction-to-the-mechanics-of- incompressible-fluids-michel-o-deville-46074298 An Introduction To The Policy Process Theories Concepts And Models Of Public Policy Making 5th Edition Thomas A Birkland https://ebookbell.com/product/an-introduction-to-the-policy-process- theories-concepts-and-models-of-public-policy-making-5th-edition- thomas-a-birkland-46137630
- 6. An Introduction to ATM Networks
- 7. This page intentionally left blank
- 8. An Introduction to ATM Networks Harry G. Perros NC State University, Raleigh, USA JOHN WILEY & SONS, LTD Chichester • New York • Weinheim • Brisbane • Singapore • Toronto
- 9. 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.
- 10. To Helen, Nick and Mikey
- 11. 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!
- 12. 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
- 13. 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 5.3 ATM Adaptation Layer 2 (AAL 2) 73 5.4 ATM Adaptation Layer 3/4 (AAL 3/4) 76 5.5 ATM Adaptation Layer 5 (AAL 5) 79 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
- 14. 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. 1 Introduction 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. 1 Introduction 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
- 15. 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
- 16. 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
- 17. 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.
- 18. 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
- 19. This page intentionally left blank
- 20. 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
- 21. 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
- 22. 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
- 23. 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
- 24. 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
- 25. This page intentionally left blank
- 26. Part 1 Introduction and Background
- 27. This page intentionally left blank
- 28. 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
- 29. 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.
- 30. 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
- 31. 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.
- 32. 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.
- 33. 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
- 34. 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.
- 35. This page intentionally left blank
- 36. 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
- 37. 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.
- 38. 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
- 39. 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,
- 40. 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
- 41. 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.
- 42. 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.
- 43. 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.
- 44. 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.
- 45. 20 BASIC CONCEPTS FROM COMPUTER NETWORKS Figure 2.9 The control field ofthe HDLC frame. N (S): send sequence N (R): receive sequence 5: supervisory function bits M: unnumbered function bits P/F: poll/finalbit. During a typical exchange of information between two stations, say A and B, both stations receive and send data. This means that there are two separate ARQ mechanisms, one for the data sent from A to B and another for the data sent from B to A. The fields N(R) and N(S)in the I-frame are used to carry information for both the ARQ mechanisms piggy-backed on the frames carrying data. N(R) is used by station A to indicate to station B the current status of the ARQ from B to A, and N(S) is used by station A to indicate the sequence number of the frame that it is transmitting to B. S-frames are used when no I-frames are exchanged, and also to carry supplementary control information. The information field is only present in the I-frames and in some U-frames. The FCS is calculated using a 16-bit CRC. A 32-bit CRC is optional. 2.5 SYNCHRONOUS TIME DIVISION MULTIPLEXING (TDM) Time division multiplexing permits a data link to be utilized by many sender/receiver pairs, as shown in Figure 2.10. A multiplexer combines the digital signals from N incoming links into a single composite digital signal, which is transmitted to the demultiplexer over a link. The demultiplexer breaks out the composite signal into the N individual digital signals and distributes them to their corresponding output links. In the multiplexer, there Figure 2.10 Synchronous Time Division Multiplexing (TDM).
- 46. SYNCHRONOUS TIME DIVISION MULTIPLEXING (TDM) 21 is a small buffer for each input link that holds incoming data. The N buffers are scanned sequentially and each buffer is emptied out fast enough before new data arrives. The transmission of the multiplexed signal between the multiplexer and the demulti- plexer is organized into frames. Each frame contains a fixed number of slots, and each slot is pre-assigned to a specific input link. The duration of a slot is either a bit or a byte. If the buffer of an input link has no data, then its associated slot is transmitted empty. The data rate of the link between the multiplexer and the demultiplexer that carries the multiplexed data streams is at least equal to the sum of the data rates of the incoming links. A slot dedicated to an input link repeats continuously frame after frame, and it is called a channel. TDM is used in the telephone system. The voice analog signals are digitized at the end office using the Pulse Code Modulation (PCM)technique. That is, the voice signal is sampled 8000 times per second, or every 125 us, and the amplitude of the signal is approximated by a 7- or an 8-bit number. At the destination end office, the original voice signal is reconstructed from these samples. As a consequence of this sampling mechanism, most time intervals within the telephone system are multiples of 125 us. The standard that specifies how to multiplex several voice calls onto a single connection is known as the digital signal level standard, or the DS standard. This is a generic digital standard, and it is independent of the medium over which it is transmitted. The DS standard specifies a hierarchy of different data rates, as shown in Table 2.1. The nomenclature of this hierarchy is DS followed by the level of multiplexing. For instance, DS-1 multiplexes 24 voice channels, and it has a data rate of 1.544 Mbps. The higher levels in the hierarchy are integer multiples of the DS-1data rate. This hierarchy is known as the Plesiochronous Digital Hierarchy (PDH). Plesiochronous means frame synchronous (from the Greek word plesio, which means frame). The DS standard is a North American standard, and it is not the same as the inter- national hierarchy standardized by ITU-T. Table 2.2 gives the international hierarchy, which consists of different levels of multiplexing. For instance, level-1 multiplexes 30 voice channels, and it has a data rate of 2.048 Mbps. As in the DS standard, the higher levels are integer multiples of the level-1 data rate. The digital signal is carried over a carrier system, or simply a carrier. A carrier consists of a transmission component, an interface component, and a termination component. The T carrier system is used in North America to carry the DS signal, and the E carrier system is used to carry the international digital hierarchy. Tl carries the DS-1signal, T2 the DS-2 signal, T3 the DS-3signal, and so on. Similarly, El carries the level-1 signal, E2 carries the level-2 signal, and so on. Typically, the T and DS nomenclatures are used Table 2.1 The North American Hierarchy. Digital signal number Voice channels Data Rate (Mbps) DS-1 24 1.544 DS-1C 48 3.152 DS-2 96 6.312 DS-3 672 44.736 DS-4 4032 274.176
- 47. 22 BASIC CONCEPTS FROM COMPUTER NETWORKS Table 2.2 The international (ITU-T) hierarchy. Level number Voice channels Data Rate (Mbps) 1 30 2.048 2 120 8 . 4 4 8 3 480 34.368 4 1920 1 3 9 . 2 6 4 5 7680 565.148 Figure 2.11 The DS-1 format. interchangeably. For instance, one does not distinguish between a Tl line and the DS-1 signal. The same applies for the international hierarchy. The DS-1 format, shown in Figure 2.11, consists of 24 8-bit slots and a 1-bit slot for frame synchronization. On the 1-bit slot channel, the frame synchronization pattern 1010101... is transmitted. Each of the 24 slots carries a single voice. For five successive frames, an 8-bit PCM sample is used. In the sixth frame, a 7-bit sample is used, and the 8th extra bit is used for signaling. The total transmission rate of the DS-1 format is 24 x 8 + 1 = 193 bits per 125 us, corresponding to 1.544 Mbps, with each voice channel carrying a 64 Kbps voice. The DS-1 format can be also used to carry data. In this case, 23 8-bit slots are used for data, and the remaining slot is used for control and frame synchronization. Each data slot carries 7 bits of data, amounting to a channel of 56 Kbps. The extra bit per slot is used for control. In the international hierarchy, the level 1 format for voice consists of 32 8-bit slots, resulting in a total transmission rate of 2.048 Mbps. Of these slots, 30 are used for voice, and the remaining two are used for synchronization and control. 2.6 THE LOGICAL LINK CONTROL (LLC) LAYER A Local Area Network (LAN) or a Metropolitan Area Network (MAN) consists of a transmission medium which is shared by all the stations that are attached to it. Access to the transmission medium is achieved through a Medium Access Control (MAC) protocol. The IEEE LAN/MAN Standards Committee has produced several standards for local and metropolitan area networks, such as the IEEE 802.3 standard for Ethernet, the IEEE 802.4 standard for the token bus, and the IEEE 802.5 standard for the token ring. The Logical Link Control (LLC) protocol was defined in the IEEE 802.2 standard, and it runs over several different MACs.
- 48. THE LOGICAL LINK CONTROL (LLC) LAYER 23 Figure 2.12 The OSI stack for LANs/MANs. The OSI stack for stations communicating over the same LAN or a MAN is shown in Figure 2.12. The data link layer in the OSI reference model corresponds to the LLC and MAC layers. As can be seen, the networking layer is not present. Typically, layer 3 carries out functions, such as routing, addressing, flow control and error control, over a sequence of links. In a LAN or a MAN, however, there is no need for routing when transmitting between two stations which are attached to the same shared medium. The other functions of layer 3 are performed by the LLC. This considerably simplifies the OSI stack. LLC is concerned with the transmission of link-level PDUs between two stations. Addressing in LLC is achieved by specifying the source and destination LLC users. An LLC user is typically a high-level protocol or a network management function. An LLC user address is referred to as a Service Access Point (SAP). The following services are provided by LLC: • Unacknowledged connectionless service: this is a datagram type of service. It does not involve any flow or error control, and the delivery of data is not guaranteed. • Connection-mode service: this is similar to the service offered by X.25. A logical connection is first set-up between two users before any exchange of data takes place. Flow and error control is provided. • Acknowledged connectionless service: this is a service which is a cross between the above two services. Datagrams are acknowledged as in the connectionless mode service, but a logical connection is not set-up. LLC is modeled after HDLC. It makes use of the asynchronous, balanced mode of operation of HDLC in order to support the connection-mode service. The unacknowledged connectionless service is supported using the unnumbered information PDU, and the acknowledged connectionless service is supported using two new unnumbered PDUs. The LLC and MAC encapsulation is shown in Figure 2.13. The LLC header contains the following fields. I/G is a 1-bit field indicating whether the destination address is an individual address or a group address. DSAP and SSAP are 7-bit fields indicating the destination and source Service Access Points (SAP). C/R is a 1-bit field indicating whether the frame is a command or response frame. The LLC control field is identical to that of the HDLC with extended sequence numbers. The MAC header contains a MAC control field, the Destination Address (DA) and the Source Address (SA), and the MAC trailer carries the FCS value. The address of a station is the physical attachment point on the LAN.
- 49. 24 BASIC CONCEPTS FROM COMPUTER NETWORKS Figure 2.13 LLC and MAC encapsulation. 2.7 NETWORK ACCESS PROTOCOL X.25 X.25 was originally approved by ITU-T in 1976to provide an interface between public packet-switched networks and their customers. Since then, it has been revised several times. It has been widely used, and it has also been employed for packet switching in ISDN. The X.25standard specifies only the interface between a user's machine, referred to as the Data Terminal Equipment (DTE), and the node in the packet-switched network to which it is attached, referred to as the Data Communication Equipment (DCE), as shown in Figure 2.14.The standard is not concerned with the internal architecture of the packet-switched network. This is done deliberately so that vendors can use their own network architectures, while at the same time they are compatible with the end users. The standard specifies the first three layers of the ISO model. As shown in Figure 2.15, X.21 is the standard for the physical layer, LAP-B (a subset of HDLC) is the standard for the Figure 2.14 The X.25 interface. Figure 2.15 The X.25 suite.
- 50. NETWORK ACCESS PROTOCOL X.25 25 data link layer, and X.25 is the standard for the network layer. Below, we review some of the basic features of X.25. X.25 provides a virtual circuit service. Two types of virtual circuits are allowed: Switched Virtual Circuits (SVC) and Permanent Virtual Circuits (PVC). The following events take place in order to set-up an SVC. A pictorial view is shown in Figure 2.16. 1. The sending DTE sends a call-request packet to its DCE requesting to establish a virtual circuit to a specific DTE. The packet contains the source and destination addresses and a virtual circuit number selected by the DTE. 2. The network routes the call-request packet to the receiver's DCE, which sends an incoming-call packet to the receiving DTE. This packet has the same format as the call-request packet, but it utilizes a different virtual circuit number selected by the receiver's DTE. 3. The receiving DTE, upon receipt of the incoming-call packet, indicates acceptance of the call by sending a call-accept packet to its DTE using the virtual circuit number used in the incoming-call packet. 4. The network routes the packet to the sender's DCE, which sends a call-connected packet to the sending DTE. This packet has the same format as the call-accept packet, and it has the virtual circuit number used in the original call-request packet. 5. The sending and receiving DTEs exchange data and control packets using their respec- tive virtual circuit numbers. 6. The sending (or the receiving) DTE sends a clear-request packet to terminate the virtual circuit. Its DCE sends back a clear-confirmation packet, and forwards the clear-request packet to the destination DCE, which issues a clear-indication packet to its DTE and from which it receives a clear-confirmation packet. Several types of packets are used in X.25. The format for data packet with 3-bit and 7-bit sequence numbers is shown in Figure 2.17. Q is a 1-bit field which is user specified, Figure 2.16 Call set-up and tearing down in X.25.
- 51. 26 BASIC CONCEPTS FROM COMPUTER NETWORKS Figure 2.17 X.25 data packet formats. and D is a 1-bit field used to indicate whether the acknowledgments are local or remote. If D = 0, the acknowledgments are between the DTE and its local DCE or the network. If D = 1, the acknowledgments come from the receiver DTE. The 12-bit field obtained by combining the fields Group no and Channel no, is used to indicate the virtual circuit number, which has local significance, i.e. it is valid only between a DTE and its local DCE. M is a 1-bit field used when packet fragmentation is employed. The P(R) and P(S) contain the receive and send ARQ sequence numbers. 2.8 THE INTERNET PROTOCOL (IP) IP is part of the TCP/IP suite of protocols used in the Internet. TCP corresponds to the transport layer of the OSI model, and IP corresponds to the network layer of the OSI model. In this section, we describe the current version of IP, known as IP version 4 (IPv4). IP provides a connectionless service using packet switching with datagrams. Packets in a connectionless network, such as the IP network, are referred to as datagrams. An IP host can transmit datagrams to a destination IP host without having to set-up a connection to the destination, as in the case of X.25, frame relay and ATM networks. IP datagrams are routed through the IP network independently from each other, and in theory, they can follow different paths through the IP network. In practice, however, the IP network uses routing tables which remain fixed for a period of time. In view of this, all IP packets from a sender to a receiver typically follow the same path. These routing tables are refreshed periodically, taking into account congested links and hardware failures of routers and links. IP does not guarantee delivery of IP datagrams. In view of this, if the underlying network drops an IP datagram, IP will not be aware of that. Also, IP does not check the payload of an IP datagram for errors, but it only checks its IP header. IP will drop an IP datagram, if it finds that its header is in error. Lost or erroneous data is recovered by the destination's TCP using the selective-reject ARQ scheme described in Section 2.3. 2.8.1 The IP header An IP datagram consists of a header and a payload. The IP header is shown in Figure 2.18, and it consists of a 20-byte fixed part and an optional part which has a variable length.
- 52. THE INTERNET PROTOCOL (IP) 27 Figure 2.18 The IPv4 header. The following fields are defined in the IP header: • Version: a 4-bit field used to indicate which version of the protocol is used. • Internet Header Length (IHL): this is a 4-bit field, and it gives the length of the header in 32-bit words. The minimum header length is five 32-bit words or 20 bytes. • Type of service: this is an 8-bit field used to indicate whether the sender prefers the datagram to travel over a route with minimal delay or a route with maximal throughput. • Total length: a 16-bit field used to indicate the length of the entire datagram, i.e. header and payload. The default value for the maximum length is 65 535 bytes. • Identification: a 16-bit field used by the receiver to identify the datagram that the frag- ment belongs to. All fragments of a datagram have the same value in the identification field. • Flags: this is a 3-bit field, but only two bits are used, namely, the 'more fragments' and the 'don't fragment'. All fragments except the last one, have the 'more fragments' bit set. This information permits the receiver to know when all the fragments have arrived. The 'don't fragment' bit is used to disallow fragmentation. • Fragment offset: the 13-bit field contains an offset that points where in the original datagram this fragment belongs to. • Time to live: this is an 8-bit field that specifies in seconds how long a datagram is allowed to live in the network. The maximum lifetime is 255 s. Every router that processes the datagram must decrease this field by one second, and by several seconds if the datagram is queued in the router for a long time. This field can be seen as being similar to a hop count. When the time to live field becomes equal to zero, the datagram is discarded. This prevents a datagram from moving around in the network forever. • Protocol: this field is 8 bits long, and it specifies the next higher level protocol, such as TCP and UDP, to which the datagram should be delivered. • Header checksum: a 16-bit field used to verify whether the IP header has been correctly received. The transmitting host adds up all the 16-bit half-words of the header using 1's compliment arithmetic, assuming that the checksum field is zero. The 1's compliment of the final result is then computed and placed in the checksum field. The receiving host calculates the checksum, and if the final result is zero, then the header has been correctly received. Otherwise, the header is erroneous and the datagram is dropped.
- 53. 28 BASIC CONCEPTS FROM COMPUTER NETWORKS The checksum is recomputed at each router along the path of the datagram, since at least one field of the header (the time to live field) is changed. • Source address: a 32-bit field populated with the network and host number of the sending host. • Destination address: a 32-bit field populated with the network and host number of the destination host. The IP addressing scheme is discussed below. • Options: a variable-length field used to encode the options requested by the user, such as security, source routing, route recording, and time stamping. • Padding: a variable-length field used to make the header of the datagram an integral multiple of 32-bit words. 2.8.2 IP addresses As we saw above, IP addresses are 32-bit long. An IP address is divided into two parts, a network and a suffix. The network identifies the physical network to which the host computer is attached, and the suffix identifies the host computer itself. The size of these two fields may vary according to the class of the IP address. Specifically, five different classes of addresses have been defined, referred to as class A, B, C, D, and E, as shown in Figure 2.19. Classes A, B and C are called the primary classes because they are used for host addresses. Class D is used for multicasting, and class E is reserved for future use. The first field determines the class of the IP address, and it ranges from 1 bit for a class A address to five bits for a class E addresses. The second field gives the network address, and the third field is the suffix which gives the host address. In class A, there is a 7-bit network address and a 24-bit host address, resulting in 128 network addresses and 16777216 host addresses. In class B, there is a 14-bit network address and a 16-bit host address, resulting in 16384 network addresses and 65536 host addresses. In class C, there is a 21-bit network address and a 8-bit host address, resulting to 2097 152 network addresses and 256 host addresses. Network addresses are usually written in the dotted decimal notation. That is, each byte is written in decimal, ranging from 0 to 255. As an example, the IP address 00000111 000000100000000000000010 will be written as 7.2.0.2. Using this notation, we have that the range of class A addresses is from 1.0.0.0 to 127.255.255.255, for class Figure 2.19 The IP address classes.
- 54. THE INTERNET PROTOCOL (IP) 29 B we have a range of values from 128.0.0.0 to 191.255.255.255, and for class C we have a range of 192.0.0.0 to 233.255.255.255. Class C is very common, whereas class A is rarely used since there are only few networks with that large number of hosts. IP reserves the host address zero to denote the address of a network. For instance, in the class B address 128.32.0.0, the network field 128.32 and the suffix is 0.0. This indicates the address of the network 128.32. For broadcasting within the network, IP uses the address 128.32.255.255. IP assigns multiple IP addresses to routers, since a router is attached to multiple networks. Specifically, a router has one IP address for each network that it is attached to. An individual host connected to multiple networks also has multiple IP addresses, one for each network connection. Such a host is referred to as multihomed. Subnetting The IP address structure described above introduces a two-level hierarchy. The first level is the network address, and the second level is the host address carried in the suffix. In many cases, two levels of addressing is not enough. For instance, if we consider an organization with a B class address, then all the hosts appear to be organized into a single group, described by the network address. However, hosts within an organization are typically grouped together to form a number of different LANs. To distinguish the LANs, the suffix of the IP address is subdivided into a subnet part and a host part. Each LAN is assigned a subnet address carried in the subnet part, and a host in the LAN is assigned an address which is carried in the host part. The actual parsing of the suffix in these two sub-fields is dictated by a subnet mask. The subnet mask is only known to the routers within the network, since the subnets are not visible outside the network. This technique is known as subnetting. Classless Inter-Domain Routing (CIDR) In the early 1990s, it became apparent that the rapid expansion of the Internet would cause a depletion of IP addresses and an explosion of the routing tables. The main cause for the address depletion was the wasteful usage of class B addresses. Typically, an organization may have a class B address, but it may only have a small number of hosts, thus leaving the host address space largely unused. Also, the routing table explosion was due to the fact that a router is obliged to keep all the addresses of all the registered networks. To alleviate these two problems the Classless Inter-Domain Routing (CIDR) scheme was proposed. This scheme permits the assignment of contiguous class C addresses, and at the same time, it reduces the number of entries required in a routing table. The basic idea in CIDR is to allocate blocks of class C network addresses to each ISP. Organizations using the ISP are sub-allocated a block of 2" contiguous addresses. For instance, if an organization requires 2000 addresses, then it will be allocated a block of 2048 or 28 contiguous class C addresses. Hierarchical sub-allocation of addresses in this manner implies that clients with addresses allocated out of a given ISP will be routed via the ISP's network. This permits all these addresses to be advertised outside the ISP's network in an aggregate manner. As an example, let us assume that an ISP was allocated 131072 class C network addresses starting at 194.0.0.0. That means that the lowest network address is
- 55. Random documents with unrelated content Scribd suggests to you:
- 56. Mózes fölütötte a könyvet, abba be volt vezetve a Sverteczky által fizetett négyszáznyolcvan forint. Kinyitotta a kasszát és talált benne kétszázat. – Hogy lehet ez? – kérdi Mózes. Az asszony próbálja elriasztani. – Mit törődöl vele? Ha eddig rám bíztad, ezentúl is rám bízhatod. – Hogy lehet ez? – kérdi Mózes ismét és már kezében tartja az asszony nyakát. – Jóisten! – hebegi ez – ne bánts Mózes. – Hogy lehet ez? – kérdi emez harmadszor. Az asszony látta, hogy nem a pénzre kíváncsi, hanem valamit sejt. Ismerte emberét és tudta, hogy el van veszve. Összeszorította tehát fehér ajkát s lehorgasztotta a fejét. – Tégy velem, amit akarsz, – mondotta. Mózes ellökte, hogy nekiesett az almáriomnak, aztán nekiült az üzleti könyveknek és egész éjtszaka számolgatott. Mikor ezzel készen volt, odaállt a kis gyerek ágyacskája elé és figyelmesen nézte. Most már látta, ami eddig föl se tűnt neki: a gyerek szakasztott Sverteczky volt, ugyanaz a fekete szem, ugyanaz a kiálló áll s mikor nevetett, az arca épúgy, mintha három darabra vált volna, mint a volt hadnagyé. Az asszony az ágyban feküdt és nem mert moccanni. Megértette minden mozdulatát s mikor a gyereket vizsgálta, megértette ezt is. Most az ember hozzálépett s az asszony megkövült rémülettel nézett rá. Azt hitte, hogy meg akarja fojtani. De Katagár Mózes csak leült az ágy szélére és lerántotta lábáról a csizmát, aztán egészen levetkőzött, lefeküdt és öt perc mulva mélyen és nyugodtan aludt. Kora reggel befogatott és elhajtott, egymagában. Délre azért otthon volt s ebédhez ült, mintha mi sem történt volna. Az asszony
- 57. tudta, hogy hol járt, de sejtelme sem volt róla, mit művelt. Tán megölte? S most rajta a sor? Halálos aggodalommal kémlelte az ember torzonborz, nyugodt arcát, s a szeméből még csak azt sem látta, hogy valami a fejében járna. Mikor elvégezte az ebédjét, odament a kasszához, kinyitotta, lassan, körülményesen kikotorászott a tárcájából egy váltót és gondosan elzárta. Ekkor állt a felesége elé és mondta: – Elintéztem mindent. Tizenhétezerkétszáz forintot adtál neki, hogy fattyú anyjává tett. A pénz az enyém, a gyerek az övé. Előbb adja meg a pénzemet, aztán majd meglássuk, mi lesz a gyerekkel. Értetted? Nem értette ugyan, de bólintott a fejével és mód nélkül meg volt rettenve. De csakhamar megértette. Mózes behívta a gyereket, aki az udvaron kergette a libákat, térde közé fogta, hosszan nézte az arcát, megsimogatta vad, fekete haját s ezt mondta: – Van-e kedved velem a Lunka Negrára menni? – Most mindjárt! – kiáltott a gyerek és tapsolt örömében. – A Lunka Negrán sok mulatság akad, – folytatta Mózes – ott örökké csilingel a nyáj kolompja, mindig vannak kecskegidák, mikkel játszogathatsz, aztán két kecskét befogunk egy kis szánba és usgyi, repülünk velök le a lejtőn. – Jaj de jó lesz! – kiáltott a gyerek – papa menjünk a Lunka Negrára! – Készülj tehát – szólt az öreg és fölállt. Akkor tavasz volt, a hegyekről egyre robogtak lefelé a vadvizek. A Lunka Negra felől távoli dörgés jelezte a hótömegek összeroppanását, a nyájak most vonultak fölfelé. Az öreg Mózes összeszedte téli holmiját, összecsomagolt néhány kiló dohányt,
- 58. néhány skatulya töltést, egy doboz teát. A gyerekkel összeszedette összes holmiait, fölrakatta az egész göncöt egy kis kétkerekű szekérre, maga pedig a gyereket kézenfogva, mondta: – Most gyerünk! Rá se hederített az asszonyra, aki kővé meredve állt ott és nézte hol a gyereket, hol azt a rémületesen nyugodt embert. A gyerek vígan anyjához futott, hogy búcsúzzék, de apja visszarántotta. – Nem érünk rá – mondta és hátra se nézve, egy szót se szólva, nekiindult a havasoknak. – Mózes! – sikoltott az asszony és utánavetette magát. Mintha nem is hallotta volna, csak ment, ment a körbe tekergődző úton, föl, a havasok felé. És leszállt a nap és elmult az éjtszaka, az ember nem jön haza. Tán valami baja történt? Jöttek a parasztok a havasok aljáról: nem látták-e Katagár uramat? De igen, találkoztak vele, fölment az a Lunka Negra felé és igen vidáman játszogatott a gyerekkel. Majd egy másik: a gyerek a vállán lovagolt, úgy vitte, bizonyára fáradt volt. De hogy is lehet ilyen kis gyereket olyan magasra vinni, különösen most, mikor patak folyik az ösvényeken? Aztán nap mult nap után és Mózes nem jött vissza. Beköszöntött a nyár, a nyájak a legmagasabb legelőkre vonultak föl és Katagár Mózesről azt hallotta az asszony, hogy egy havasi kunyhóban ütött tanyát a kis Julissal, aki igen vidám. Puskával a vállán barangolja be a sziklákat és egyre teregeti le a zergéket. – Nem izent semmit? – Semmit. S az ősz is beállt és a kis oláh falu éjjel-nappal, heteken keresztül ázott a szakadatlan esőben. Sverteczkyről hírt se hallott többet. A Retyezát fehér jéglapjai eltűntek a felhők mögül, a pásztorok vissza- visszatértek már a faluba, a hegyeket egészen ellepte már a hó. Csak Katagár Mózes maradt fönt a kislánnyal, ahol most farkasok ordítanak versenyt a bömbölő széllel.
- 59. Jött a karácsony is, egészen fehéren, egészen reménytelenül. A hegyi utak járhatatlanok, most már nem is jöhetnek le, tán el is pusztultak odafönt. Ez a folytonos várás, rettegés, tépelődés, a lelkifurdalás, a gyerekért való gond, közepette a nyugodalmas, egyhangú életnek a boltban, a számadások között, csontig lesorvasztották a szegény asszonyt. Imádkozott, hogy ura jöjjön vissza és lőjje le. Szilveszter estéjén aztán váratlanul betoppant. Úgy, mintha mi sem történt volna. A kislányt karján tartotta s az vastag nyakát átölelve, aludt. Az asszony szólni sem mert, csak úgy lihegve, vágyva nézett a gyerekre, de nem merte érinteni. Mózes rá sem nézett, csak gyöngéden levetkőztette a gyereket és lefektette. Aztán maga is levetkőzött és lefeküdt. Másnap kiállt a boltajtó elé és pipált. Csikorgó hideg volt, ott fönt a havasokban megszokta. A kis gyerek, mikor fölébredt, csodálkozva nézett körül, egészen elszokott a szülei háztól, tán nem is igen emlékezett rá. Édesanyja most, hogy az ember nem látta, rávetette magát és keserves könnyhullatás között csókolgatta. A kislány tűrte, idegenül, szinte kedvetlenül. – Elfelejtetted már anyádat, nem szereted már anyádat? – zokogott a szegény asszony. A gyerek nem felelt, csak nézett vágyva ki a kis ablakon, a szabadba. – Ugy-e jó itt? Most már itt maradsz! – szólt reménykedve, meg hogy valamit megtudjon, az asszony. – Odafönt jobb – volt a kurta felelet. – De nem engedlek, itt tartlak, te szívtelen gyermek! – kiáltott az asszony. – Apa erősebb – volt a felelet.
- 60. Délfelé az öreg Mózes nyugtalankodni kezdett. Egykedvű arca elsötétült, majd fenyegető lett. Nézte az óráját s a kezében tartogatva, kémlelte az utat. Mikor a mutató pontban tizenkettőre ért, Katagár Mózes bement a boltba. – Jól jár az óra? Ez volt első szava a feleségéhez. – Jól jár. – Megnézte, ott még egy negyedóra hiányzott tizenkettőre. Ekkor száncsörgés hallatszott, a bolt előtt állt meg s belépett Sverteczky. Egészen lefogyva, arcán a hajszolt vad fáradtsága és a kísértettől remegő kutya félelme. Egy pillanatra találkozott a szeme az asszonyéval, de akkor már ott állt Katagár Mózes. – Megvan? – kérdezte. – Megvan – felelte amaz halkan. Bementek a belső szobába. Ott Sverteczky egy váltót adott át és néhány nagy bankót. Katagár Mózes gondosan megolvasta a pénzt, elzárta, aztán kivette és átadta Sverteczkynek a régi váltót. – Mához egy esztendőre, – mondotta és elfordult. – Mózes, – szólt az most – valamit szeretnék kérdezni. – Az üzletre tartozik? – Arra. – Nos? – Mi lesz, ha véletlenül meghalnék? – Az a ti bajotok, gondod legyen rá, hogy ne halj meg. Én is vigyázok magamra. Ha vele akarsz beszélni – nem bánom.
- 61. Nem bánta, sőt maga ment ki a boltba, míg azok ketten beszélgettek. A gyerek nem tágított oldala mellől. Ekkor tudta csak meg az asszony az egész szörnyűséget. A megállapodás, azaz Katagár Mózes határozata az volt, hogy míg az egész pénzt le nem fizette, addig a gyereket magánál tartja. Ha pedig bármely oknál fogva elmarad a törlesztés, megöli a gyereket. S ezzel a rémítéssel kergette végig az országon azt az embert, pénzt szerezni. Kényszerítette, hogy vigyázzon az egészségére és éljen rosszabbul, mint a kutya és takarékoskodjon, mint a fukar. Ha gályarabságra ítéli, nem olyan szörnyű a büntetés. S míg az embert végigkergeti az országon, az asszonyt egyhelyben, egymagában fogva tartja a boltban. Ez húsz esztendeig fog tartani, mert annyihoz kötötte magát az öreg Katagár. Még két napig lent maradt s beszerezte a szükségleteit. Aztán hátra se nézve, kézen fogta a gyereket és elindult ismét a vadonba. A gyerek se nézett hátra, egészen olyan volt, mint az az ember, akit apjának tekintett. És év év után megismétlődött ugyanaz a jelenet. Minden esztendő sorvadtabban, megtörtebben, megvénültebben hozta össze a hajdani szerelmes párt, és erősebben, duzzadó egészségben piros orcával és büszke fejhordással a megcsalt férjet. S a gyerek is megbarnult, szinte megnémult, de erős, durva szépségű női óriássá fejlődött. A pásztorok között regék jártak róla, a havasok odafönt százszorosan visszaverték vidám kacagását, harsogó danáját és a puska ropogott a kezében s úgy utánairamodott a zergéknek, hogy szinte elevenen fogta el. Itt lent néma volt és csöndes, rá se nézett senkire, csak az apján csüngött a szeme, rajongó szeretettel és boldogsággal. Mikor a leány tizenöt éves lett, Katagár Mózes nem vitte többé a havasokba. Ellenben összeszedett egy nagy csomó pénzt s elment vele Pestre. Az ember és a leány nagyon jól értették egymást, odafönt annyira összeszoktak, hogy a gondolatuk is találkozott s
- 62. amit az öreg akart, az mindig jól esett a lánynak. Szép lakást rendezett be Pesten és járatta a házhoz a különböző mestereket. A lány tanult nyelveket, tanult zongorázni, festeni és táncolni. Olyan erős és egészséges volt, amivé csak tízévi merőben fizikai élet tehet, istenverte magánosságban, a legzordonabb környezetben. És ezt a megedzett anyagot most megcsiszolgatta kulturával. A durva arc, a szinte fekete kéz megtisztult, megfinomult anélkül, hogy elpuhult volna. Három év multán Katagár Juliska olyan szépség volt, amilyen Magyarországon még nem termett. Sverteczky pedig egészen összeaszott, most már csak mankón tudott járni. De hűségesen, mintha kísértet kergette volna, minden újév napján ott volt a kis oláh faluban és hozta a törlesztést meg az új váltót. És látva azt az asszonyt, akivel vétkezett, félig megfehéredve, félig megkopaszodva, egészen belefásulva ebbe a rettenetes életbe, amint gépiesen méri a pálinkát, a lisztet, a cukrot, petróleumot, kocsikenőcsöt, szöget, sarlót, kaszát, minden évben kegyelemért könyörgött, de Mózes hajthatatlan maradt. – Az utolsó garasig, az utolsó napig – mondotta – a gyerek a kezemben van és tudod, hogy én megteszem, amit mondok. Juliska huszonegy éves korában férjhez ment. Igen előkelő, igen gazdag ember vette el. De Katagár Mózes föltételül tűzte, hogy míg a leánya huszonhat esztendős nem lesz, addig az ő házánál kell maradnia. Fényes, nagy házat vitt és époly könnyűséggel merült bele a nagyúri életbe, mint annak idején a havasiba. Ezt az embert semmi se hozza ki a sodrából. Végre elkövetkezett a huszadik év és Sverteczky úgyszólván a sír széléről támolygott a kis faluba s olvasta le az utolsó részletet. – Most rendben vagyunk – szólt az öreg Mózes. – Most már meghalhatok – szólt Sverteczky. – Most visszakaphatod a lányodat, – szólt Katagár Mózes – de becsületes módon, hogy kára ne legyen belőle.
- 63. – Hogyan? – hebegte Sverteczky. – Azt majd este megtudod. Délben történt ez a beszélgetés és estére az öreg Katagár Mózes halva volt. Főbe lőtte magát, Néhány sort hagyott hátra a feleségének: meg fogtok most esküdni egymással, hogy becsületes szülei lehessetek a lányotoknak.
- 64. KÉT ÁGY. Szép vagy, én édesem, ezidőszerint egyetlenem, de a feleségem sokkal szebb tenálad. Mi a te buja húsosságod az ő ártatlan nádszáltermetéhez képest. A te beszédes, csókos, kipárnázott ajkad ahhoz a keskeny, rózsaszín vonalhoz képest, mely beszédesen és mosolyogva minden szónál, minden hangulatnál másfelé kígyózik s mikor ajkammal megközelíteném, hirtelen csúcsos, piros tölcsérré csucsorodik, olyan, mint egy kis nyíló kráter, mely magába szívja a lávát! Soha úgy nem szerettelek és nem foglak soha úgy szeretni, mint őt, akit veled megcsalok, akit elhagytam, mivel nem engedtek tőle elválni. Az emberek nagyon sokat spekulálnak és csóválgatják a fejüket, hogy miért jár az örök hűség fogadalma nyomán a hitszegés és a kiábrándulás. Évszázak óta törik rajta a fejüket, hogy miért szeretlek téged jobban, mint a feleségemet, pedig téged nem is szeretlek. A férfi csapodár teremtés – de a két év alatt, hogy »boldog« házasságban éltem vele, csapodár természetem minden szép alak felé hajlott, csak feléje nem. Hogy megvan, hogy birtokomban van s csak a nehézség, meg a tilalom ingerli az embert? Te is megvagy és semmi nehézségembe nem került a meghódításod. Leírom az én történettelen regényemet azok okulására, akik hasonló bajban vannak s megvan a komoly szándékuk a bajtól való menekülésre. A főbaj a két ágy, a közös hálószobában. Amikor a boldog mamák kiválogatják a bútorosnál szerencsés gyermekeik számára a családi ágyakat, akkor koporsót vásárolnak gyermekük szerelmének és boldogságának. Mikor jegyes korunkban, két nappal az esküvő előtt megnézvén lakásunkat, – nekünk semmi beleszólásunk nem volt, mivel a menyasszony szülei fizették a bútorokat – beléptünk a hálószobába, bennem reszketett minden ideg és ész nélkül magamhoz kaptam a remegő szép leányt.
- 65. – A hálószoba – suttogtam és majd megrepedt a fejem. És kémleltem az ő szép, finom gyermekarcát, melyet gyönge pirosság vont be. S az első nyilallás akkor támadt a szívem tájékán. Oda tudott lépni az ágyhoz és annyi műértéssel megtapogatta a paplan selymét, végigsimította a habos párnákat és gyönyörködött a tenyérnyi monogrammban. – Ez lesz az enyém, – mondotta, a fal felé álló ágyra mutatva – nem tudok az ablak felé aludni. Különös volt. Én nem tudtam megbarátkozni a valóság gondolatával, hogy ezzel a liliomszállal egy szobában fogok aludni. Pedig a kalandok emléke egész árbocerdőként meredezik mögöttem ég felé. Ő a szende, a tiszta, az ártatlan, úgy beszélt róla, mintha azt mondaná: holnapután hetivásár lesz. – Kedvesem, – mondtam már akkor – ha rajtam állna, külön hálószobát rendeznék be neked. – Miért? – kiáltott szinte kétségbeesve – hát nem szeretsz? Zavarban voltam s akkor a magam vágya ellen is szóltam. Csakugyan, miért? A két ágy egymás mellett örökkévaló ölelést jelent. Melyik szerelmes térne ki ez üdvösség elől? – Mert szeretlek – rebegtem őszinte érzéssel és gondolattal – és szent vagy te nékem, mint feleségem is és nem akarom, hogy a tulajdonommá válj, mikor feleségemmé teszlek. Ennek sok értelme nincs és ő nem is értette meg. Csak sírt és alig tudtam nagynehezen megbékíteni. – Hát ez nem tartozik a házassághoz? – kérdezte panaszos hangon. – Hogyne, hogyne. Bizonyos körökben nincs úgy, de a polgári osztálynál ez a házassághoz tartozik. Sőt a szegény embereknél az tartozik a házassághoz, hogy a két ágy egy ágy legyen.
- 66. – Ez förtelmes – kiáltott föl és elfordult. A jó isten érti, miért volna ez förtelmesebb, mint a két ágy, de iszonyodása mégis jól esett. Az olyan romlott ember, mint jómagam, szörnyen naiv az ártatlan naiv asszonykához képest. Könnyelmű életében könnyelmű nők között gyüjtvén emberismereteit, a rossz és az ártatlan nőt annyira különböző két isten teremtményének véli, hogy szinte belekábul, mikor megismeri a magától értetődő igazságot, hogy a tiszta nő is: nő. A mennyország is, a pokol is a hitnek egyívású pólusai, mind a kettő a jó istennek világfenntartó eleme. De hát melyik férfi élte végig viharos ifjúságát romlatlan teremtések közepette? Melyik nem jár úgy, mint én, hogy vágyódik a szennyből, az olcsóságból, a léhaságból a piruló tisztaság után, ahol a csók szentség, és a nő, mikor karjába zárja a férfit, esdeklő pillantással, akarata nélkül, elbűvölve és kiszolgáltatva rogy az ölelő szerelem forgatagába? Egy tiszta, szemérmes nő oldalán – ez a képzet a maga költői varázsával megutáltatta velem eddigi életemet. És mesék meséje a képzeletnek valóra válása és poklok pokla, az éden kertjének kínszenvedéses elvesztése, mikor az ember tapasztalja hogy ez a valóság csak egyszer valóság, aztán – hajrá! vágyódhatsz ismét bűvös, remegő szemérmetesség után. Az én szegény kis feleségem nem tehet róla hogy máskép képzeltem a tisztaságot. Mire hazaértünk a nászútról, én megtanultam pirulni, ő elfelejtette. A szerelem új világa nyilt neki és megrészegítette. Az odaadás gyönyörűségébe beletorkollott a fiatal asszonynál a becsületes, őszinte átérzése a frázisnak: férj és felesége egy test, egy lélek. Jól értem, ez csupa tisztaság, de az ördögbe, mikor voltam én szerelmes a magam testébe? – Minek szégyeljem magamat? – mondotta az angyal meggyőződésével – hiszen egyek vagyunk! Persze, persze. De amikor a férjet éppen az vonzza és hevíti szerelemre, amitől a nő iszonyodva menekül: az idegenség? Ti, okos
- 67. férfiismerők, tudjátok azt és azzal tartjátok fogva a férfi csapodárságát, hogy egy testetekből, egy szépségetekből minden napon mást csináltok. Egy új toilette: új asszony, új ingerrel, új szerelemmel a régi, a megismert helyében. De mit ért ehhez egy erényben növekedett polgári asszonyka? A szerelem minden igézetét beleöli a jogba és megengedettségbe. De mikor éppen a jog és a megengedettség kövezi ki az utat a megszokottság és meguntság felé! Mit ád az ilyen asszony külsőségekre – az ura előtt? Hétszámra ugyanabba a pongyolába bujik. A női fehérnemű, csipkés, habos tisztaságában kész bűvölet, mint a madár tolla, a virág színpompája. Nekem látnom kell, mint válik a csipkés sejtelemből szennyes. A nő üde arca a szépség igéretföldje – de a férj előtt minek mesterkedni? Nekem látnom kell, hogy reggelre ébredve az üde, szép arcot mint lepte el az éjtszakai alvás fakósága, mint duzzasztja a bőr zsírosodása, mint dohosítja a rászáradt verejték. Istenem, hát vak ez az asszony? Erőnek erejével széjjel akarja tépni az illuziókat? Nem tudja-e, hogy minden nő egyforma, ha az illuzió nem tesz közöttük különbséget? És ez az átkozott, ez a végzetes egyszerűség, a könnyű, kedves, formát adó ruhácska alatt a praktikus, a durva alsó! A csíkos alsószoknya, meg a vastag barchet-nadrág, mely térden alul csúszik. S látnom kell a feleségemet, aki olyan mint a péklegény a műhelyben s szerelmes vággyal vagyok köteles a péklegény mozdulatait követni s rettegve gondolok a pillanatra: no most a kebledre fog borulni. Belenéznék gyönyörű tószín szemébe, attól elfelejtenék barchetet, péklegényt, meg mindent, de szép szemét lehúnyja az ártatlan teremtés és én nem tudok szabadulni a péklegény-hasonlattól. Az isten szerelmére, miért nem adják a leányokat egy tapasztalt kokotthoz iskolába? Az életnek boldogsága attól függ, hogy vetekedni tudjon veletek és büszkén és gőgösen ragaszkodnak ügyetlenségükhöz, sírjukat ásó tudatlanságukhoz. Nem az a fő, hogy a gyermeket nem a gólya hozza s azok az ártatlan teremtések azt
- 68. hiszik, hogy most már asszonyok, amikor megtudják, hogy a gólya kelepelő szárnyas, aki csak békákat és kígyókat fogdos. Annyi évszázadon keresztül panaszkodnak az asszonyok, hogy mily komiszak a férfiak, mennyit kell tőlük szenvedniök s mégsem jut eszükbe, hogy valamit kitaláljanak e komiszság ellen, valamit, ami már régen megvan s aminek az elhanyagolása teremti meg a férfikomiszságot és a női boldogtalanságot. Ha pedig teljes nyers műveletlenségben akarnak élni, hát kérdezlek, nem éppen az ő érdekük-e, hogy minden fogyatékosságuknak ne legyen tanuja az az ember, aki előtt tökéletesnek akarnak és kell is látszaniok? S még egy csekélység. Jóérzésű ember nem alhatik másodmagával. Parasztnak és munkásnak való ez, aki mihelyest behúnyja szemét, már hortyog is. A magunkfajta ember többé-kevésbbé mesterségesen altatja el magát. Az egyik olvas, a másik mozdulatlanul fekszik. Már most én olvasnék, de kedves nőmet bántja a lámpa fénye. Ő hamarosan elalszik, én a sötétben fetrengek. S annyi neme, faja van az alvó embernek. Fordul, csapja a takaróját, nyög, megrezzenti az ágyat, hát miért ne alhasson nyugodtan és úgy, ahogy megszokta, az a szerencsétlen ember, aki férj? S mi ennek a következése? Ha nagyon álmos voltam már és egyszer kedvemre aludni akartam, vasútra ültem és elutaztam képzelt üzleti ügyek miatt. Ez volt az első stáció. A második előtt még egyszer meghökkentem, mert én igazi férj akartam lenni, aki hűséget tart a feleségének. Minden asszony, aki nem velem alszik egy szobában, őrült szerelmet keltett bennem. Mikor fogyni éreztem erőmet s a kísértés már-már megejtett, komolyan beszéltem a feleségemmel. – Kedvesem, nekünk mégis jobb volna, ha külön hálószobája volna mindegyikünknek. Egymásba nyíló, az ajtó akár nyitva is maradhat, de ajtó legyen közöttünk, melyet esetleg be lehessen csukni.
- 69. – Hát nem szeretsz már? – kiáltotta kétségbeesetten és ettől valóságos dühroham fogott el, melyet alig tudtam legyűrni. – Éppen, mert, szeretlek – szólottam – kedvesem, vannak dolgok, miknél csak egy csúnyább van a világon: az elmagyarázásuk. Engedd el a magyarázatot és tégy a kedvem szerint vakon, de bízva. Hát hogy asszony olyasmit tegyen, amit nem ért, arra nincs eset. Ekkor azt ajánlotta nekem, hogy váljunk el. Hát ez is furcsa dolog. Egészen elválni: ok nélkül, az természetes, de egy ajtórésnyire távolodni egymástól: soha! Az imént szavamba vágtál, ellenem vetvén: hát a lélek? a lelki közösség? Édes barátnőm, nagyon ritkán van férj és feleség között lelki közösség. Nálunk például nincsen. Megszerettem a nőmet szépségénél fogva, elvettem, mert minden egyéb külső föltétel megfelelt. Nem vizsgáltam meg: mit tud, s fiatal leányról senki sem tudhatja, hogy milyen, mert abban a tisztelő, szűzies távolságban lappang a megtévesztő hamisság. Sőt ha utolsó gondolatáig ismerem is a leányt, nem tudhatom, milyen lesz, mint asszony. Az asszony abban a pillanatban születik, amelyben először karjába zárja a férfi, addig nem élt. S az új asszony éppen nem ugyanolyan, mint a leány volt, sőt a legritkább esetben olyan. A szerelem olyan minden gondolatát és képzetét átalakító fölfedezés, mely után kő kövön nem marad a leány világában. Hát mi közöm nekem a nő lelkéhez? Egyetlen közöm volna hozzá, de az éppen testi szépsége és kívánatossága útján szövődik. Aki elmámorosodik a nő karjaiban, annak édes és szép a lélek is és a csókokkal gondolatokat és lelket cserélnek. Az én nőmmel nem volt semmi lelki közösségem, különben megértett volna, különben lehetséges lett volna megértetnem magamat. De mikor arra határoztam magamat, hogy részletekbe menjünk, megint csak fölkiáltott: – Hát utálsz engem? S azóta sportszerüen mindig azzal kínzott:
- 70. – Tudom, hogy utálatos vagyok, tudom, hogy undorodol tőlem. De nem kötelessége-e az asszonynak, ha ezt csak félig sejti is, tenni róla, hogy férjének ne legyen oka ilyen förtelemre? Ahelyett azonban ő a tényt bűnnek tette meg: férjnek, aki szereti a feleségét, mindenképen szépnek kell találnia asszonyát, ha nem találja annak, akkor nem szereti. S most én vagyok a gonosz, a hűtelen, a csélcsap. Egy új alsószoknyát ha magára vesz, kétszáz pengővel többet költ ruhára és valami ízlést, luxust fejt ki egyebekben: mi boldogan élnénk ma is. Vagy legalább ezt az egyik ágyat tétette volna át a másik szobába. Nem, jött az anyós is és kijelentette, hogy ő harminc éven át aludt az ura mellett, azért mégis tisztességes ember az ura. No hát, én nem vagyok az. Hazudni nem akarok, csalni még kevésbbé, hát jó, legyen: váljunk el. Ilyen válópör nem folyt még bíróság előtt. Neki is csak az az oka volt, hogy én nem szeretem már. Nekem külön hálószoba kellett. A bíró tárgyalás közben azt mondta: nevetséges. A majom. Nincs súlyosabb ok a válásra, mint a közös hálószoba. És nem választottak el. Bírói utasításra, törvényes szankció mellett élem most a feleségem oldalán a legényéletet. Pocsolya a családi tűzhely, sóhaj és könny a családi élet tartalma. Sajnálom szegénykét, de nem tehetek róla. Ő akarta voltaképen. És most még egyszerűbben jár, mint tisztességes, elhanyagolt asszonyhoz illik s napról napra jobban sajnálom és jobban iszonyodom tőle. De a családi két ágyhoz ragaszkodik. Mert ez a szerelmi boldogság és a tisztes családi élet szimboluma és a nagyanyja is így élt. Én pedig már nem is kérdezem: a szimbolum a fődolog-e, vagy az, amit szimbolizál?
- 71. VAK ASSZONY. I. Az ismerősök, jóbarátok mosolyogva suttogták, hogy Kludiéknál az úr fogadja föl a cselédeket. Megesett, hogy a különben nem nagyon serény ember pusztán szobaleányért utazott föl Budapestre és hozott is magával egyet, olyat, hogy a szeme is káprázott annak, aki ránézett. A gyönyörű teremtés úgy is viselte magát, mint akit gyorsvonaton hoztak, külön kupéban. Figyelmes szemű gazdasszonyok legalább észrevették, hogy a Juliskának fehér kötője és takaros klottruhája mindig tisztább, mint a bútorok; ablakmosáson, padlósúroláson pedig sohasem érték. Az érdeklődő asszonyságok – nincs ebben semmi, ha asszonyok cselédről beszélgetnek – puhatolództak is Ágnes asszonynál: hogyan van megelégedve az új szobaleánnyal. Elállt a szájuk, mikor azt az egykedvű választ kapták: – Majd elkergeti az uram, ha nem lesz vele megelégedve. Nekem mindegy, akármilyen a szobaleány, mert én csak a szakácsnénak veszem hasznát. Nevettek is rajta, szörnyűködtek is. Hát vak ez az asszony? Vagy nincs szíve? Tudvalevő, hogy az Ágnes szülői a végsőig ellenezték ezt a házasságot, mert féltették gyermeküket a Kludi Gábor csapodár hirétől. Csakhogy Ágnes kijelentette, inkább nekimegy a Marosnak, semhogy Gáborról lemondjon. Tehát szerelmi házasság volt – legalább az ő részéről. S most az urára bízza, mikor kergesse el a szobalányt. Igaz, Ágnes asszonynak egyéb dolga is van, mint leskelődni a cselédszoba ajtaján. Öt éve, hogy férjhez ment és négy gyerek sivalkodik a háznál. Sőt mintha az ötödik is útban volna már.
- 72. Szegény, sajnálni való teremtés különben, aztán úgy is lehet, hogy ez a túlságos fogékonyság az anyai hivatás iránt, okozója minden bajnak. Már az első gyerekágy után a karcsú, nyúlánk Ágnes komlósrúddá soványodott, finom elefántcsontszínű arca fakóra vált, sőt ragyogó szőke hajából is kiveszett a csillogás. Nem csoda, ha a férj, aki szerette a szép leányt, nem találja meg szerelme tárgyát a csúnya asszonyban. Hogyan van, mint van? Én nem tudok többet a szomszéd asszonyoknál. A békesség különben mintaszerű a háznál. Még a cselédleány sem hallott soha semmi veszekedést férj és feleség között. Ellenkezőleg, nagyon sokan tanui voltak a kiváló gyöngédségnek, melyben Kludi Gábor feleségét részesíti. No meg a minden esztendőben kelepelő gólya is jelenti, hogy abban a házban csönd és békesség uralkodik. Hát éppen azért támadt Ágnes asszony felől az a különös, félig gúnyos, félig lenéző hír, hogy asszony, akit az ura a szobaleányával aláz meg és nem tud róla! Lehetséges ilyesmi? S ha igen, hol a hiba: a fejében-e, vagy a szívében? Ágnes asszony különben maga a mosolygó türelem. A gyerekek mind olyanok, mintha egyetlenek volnának. Folyton rajta lógnak, agyoncsigázzák. Hónapok múlnak el és Ágnes asszony nem kerül ki az utcára. Pedig jókedvű leány volt, táncos, kacagós kedvű. Most valami különös mosoly lebeg az ajkán, mely félig lomhaságot, félig fásultságot jelent. Ha vendég van a háznál, alig látni. Rendszerint a leves után bocsánatot kér: a gyerekekhez kell mennie. Este, ha férfitársaság gyülekezik, Ágnes asszony nagyon korán visszavonul és urának sohasem kell búcsúznia tőle, ha a jókedvű társasággal átvonul a kaszinóba, ahonnan csak hajnalhasadáskor vetődik haza. Amennyire az asszonyok lenézik Ágnest, a férfiak annyira irígylik Gábort. – Ilyen asszony kellene! Boldog ember, aki sohase hallott kukli- prédikációt!
- 73. Kludi Gábor ilyenkor nem szerette folytatni ezt a témát. Bizony néha, mintha jól esett volna neki egy kis szemrehányás az asszony részéről. A nyárra vendégük érkezett. Rendes Gézáné, az Ágnes unokanénje. Mintegy harmincesztendős asszonyka, pezsgő temperamentummal, erős, szélesvállú, pirosarcú pesti asszony, aki nyaralás ürügye alatt itt, ezen az elmaradt erdélyi vidéken ragyogtatta káprázatos toalettjeit. Ennek a látogatásnak az volt a következménye, hogy a szép pesti asszony egy hét folytán összepörölt a szőke szobaleánnyal, akinek nem igen imponált a selyem pongyola s ugyancsak visszanyelvelt. Ágnes asszony különös fásult mosolygásával hallgatta a pörlekedést, de nem szólt. A szép asszony gyönyörű fekete hajába markolt rózsás körmeivel, úgy támadt reá: – És te eltűröd, hogy a cseléd így bánjék vendégeddel? Ágnes asszony, aki ölében tartotta a legkisebbiket, míg a két nagyobbik gyerek kétoldalt iparkodott, ki a hátára, ki a nyakába mászni, szelíden válaszolt: – Majd ha Gábor hazajön, rendet csinál. Rendesné egy pillanatra meghökkent, gyilkos szemmel mérte végig a szobalányt, aki kihívó mosollyal állta tekintetét. Mintha ebben a pillanatban megértették volna egymást s nem mint uriasszony cseléddel, hanem mint asszony az asszonnyal álltak volna egymással szemben. Kludi Gábor hazajött és rendet csinált. Azonnal elcsapta a szép szobaleányt. – Rögtön pakkoljon és kotródjék – rivallt rá. A szép leány dühre fakadt.
- 74. – Jó, megyek, de hát elébb a nagyságával fogok beszélni. Meg kell neki is tudnia, mi dolgom volt nekem ebben a házban és miért kell most elmennem. Ágnes asszony szelíden, de határozottan felelt: – Velem nincs semmi beszélni valója, ha a nagyságos úr elbocsátotta, menni fog s cseléddel nem állok szóba. És elvonult a hálószobába, a ház végén, hogy ne hallja a mosdatlanságokat, miket az elbocsátott cseléd pakkolás és távozás közben torkaszakadtából kiáltott, hogy a szomszédság ablakai kinyiltak és ifjak meg öregek kaján mosolygással, durva érdeklődéssel mohón magukba szítták. Ő tehát megint nem tudott semmit, az egész város pedig arról beszélt, a szép pesti asszony miként túrta ki a szép pesti szobaleányt. Sajátságos élet folyt azután. Az idegen azt hitte volna, hogy Rendesné a ház asszonya, míg Ágnes csak amolyan fizetett, alázatos gouvernante, aki a gyerekekkel foglalkozik. Alig mult nap, hogy az eleven pesti asszony valami kirándulásra nem ment, természetesen a Kludi Gábor kíséretében. Majálisokat is rendeztek, nyári táncmulatságokat. Rendesné mindenütt a Kludi Gábor karján jelent meg. Ágnes asszonyt sohase látták. Az is megesett, hogy a szép asszony, ha már idáig jött, kíváncsi volt megtekinteni Nagyszebent. Délben indultak el útnak és csak harmadnapra érkeztek vissza. Ilyet se tűrt volna el más asszony, de hát ennek az Ágnesnek valóban nincs vére. Megfoghatatlan egykedvűsége ingerelte Rendesnét is. – Mondd csak, Ágnes, – szólt hozzá – nem vagy rám féltékeny? Ágnes vállát vonogatta. – Nem féltem az uramat, – felelt csöndesen.
- 75. – Senkitől? – Senkitől. – Annyira hiszed, hogy szeret? – Annyira hiszem, – szólt Ágnes, erősen szemébe nézve a kémlelő asszonynak, – hogy Gábornak szüksége van rám és amit én nyujtok neki, azt senki másnál még csak keresni sem fogja. – És – tudja azt ő is? – Ma még nem – felelt egyszerüen és ölébefogva a sivalkodó gyereket, kezdett neki mesélni: volt egyszer egy igen szegény leány, úgy hívták, hogy Hamupipőke. Rendesné végighallgatta a mesét, melynek hossza-vége nem volt, mert mikor Hamupipőke már a király felesége lett, a gyerek belemarkolt az édesanyja hajába s mintha csöngetyűzsinór volna, úgy rángatta: – Még ne legyen vége, akarom, hogy a mese tovább tartson. És a mese tovább tartott. A könnyelmű pesti asszonyon, mintha megindulás vett volna erőt. – Hallod Ágnes, – szólt fojtott hangon, – ha te azt hinnéd, hogy az én ittlétem árt neked, hát én elutazom. Ágnes szelíden nézett rá. – Semmi okod rá. Gábor nagyon szereti, hogy itt vagy. Arra kérnélek, hogy hosszabbítsd meg ittmaradásodat, amíg éppen lehet. Rendesné erre nem felelt semmit, csak nézte, nézte a fonnyadt asszonyt, aztán hirtelen odalépett hozzá és megölelte. Ágnes nem viszonozta ölelését, de tűrte. Bekövetkezett az aratás ideje, Gábor a tanyára vonult s nem is tért haza éjtszakára sem. Rendesné mindennap kikocsizott hozzá,
- 76. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com