91,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Serie: Wiley - IEEE
- Sprache: Englisch
The telecommunications industry has advanced in rapid, significant and unpredictable ways into the 21st century. Global Networks: Design, Engineering and Operation guides the global industry and academia even further by providing an in-depth look at the current and developing trends, as well as examining the complex issues of developing, introducing, and managing cutting-edge telecommunications technologies. The author draws upon his considerable experience in the telecommunications industry to educate engineers designing equipment and systems on the hardware and software features essential to fault tolerant operation. He describes how to design networks that are fault tolerant and global in scope; how to identify best engineering and operations practices; and examines the role of technology labs in carrier networks. Software and hardware engineering practices are covered in depth. Hardware and software designs are explained with an emphasis on application and interaction of craft and operators with equipment and systems. The author proposes that equipment, systems and network designs should be integrated with the engineering and operations teams that run them. Practice, experience and a historical background are used to describe which designs and technologies fit which network services and applications. Global Networks is a complete and thorough assessment of the communications industry today, written by an author of international renown.
Key features:
- Comprehensive treatment of the key theories and technologies associated with the design of modern communications networks, including equipment, systems and network design
- Coverage of equipment and software design, mobile networks, integration and the characteristics of large network outages
- Written in an accessible style and fully illustrated, it offers a complete and up-to-date picture of communications technologies from initial design through to application
- Includes a section on future challenges such as the Exabyte traffic growth and an assessment of the dual roles of IPV4 and IPV6
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 781
Veröffentlichungsjahr: 2012
Ähnliche
Table of Contents
Title Page
Copyright
Dedication
List of Figures
About the Author
Foreword
Preface
References
Acknowledgments
List of Acronyms
Part I: Networks
Chapter 1: Carrier Networks
1.1 Operating Global Networks
1.2 Engineering Global Networks
1.3 Network Taxonomy
1.4 Summary
References
Chapter 2: Network Systems Hardware
2.1 Models
2.2 Telco Systems Model
2.3 Modular Computing—Advanced Telecommunications Computing Architecture (AdvancedTCA TM)
2.4 Blade Center Model
2.5 Summary
References
Chapter 3: Network Systems Software
3.1 Carrier Grade Software
3.2 Defensive Programming
3.3 Managed Objects
3.4 Operational Tests and Fault Conditions
3.5 Alarms
3.6 Network System Data Management
3.7 Summary
References
Chapter 4: Service and Network Objectives
4.1 Consumer Wireline Voice
4.2 Enterprise Voice over IP Service
4.3 Technology Transitions
4.4 Summary
References
Chapter 5: Access and Aggregation Networks
5.1 Wireline Networks
5.2 Hybrid Fiber Coax (HFC) Networks
5.3 Wireless Mobile Networks
5.4 Wireless Design and Engineering
5.5 Summary
References
Chapter 6: Backbone Networks
6.1 Transport
6.2 IP Core
6.3 Backbone Design and Engineering
6.4 Summary
References
Chapter 7: Cloud Services
7.1 Competition
7.2 Defining the Cloud
7.3 Cloud Services
7.4 Summary
References
Chapter 8: Network Peering and Interconnection
8.1 Wireline Voice
8.2 SS7 Interconnection
8.3 IP Interconnection
8.4 Summary
References
Part II: Teams and Systems
Chapter 9: Engineering and Operations
9.1 Engineering
9.2 Operations
9.3 Summary
References
Chapter 10: Customer Marketing, Sales, and Care
10.1 Industry Markets
10.2 Consumer Markets
10.3 Enterprise Markets
10.4 Summary
References
Chapter 11: Fault Management
11.1 Network Management Work Groups
11.2 Systems Planes
11.3 Management Systems
11.4 Management Domains
11.5 Network Management and the Virtuous Cycle
11.6 Summary
References
Chapter 12: Support Systems
12.1 Support Systems Standards and Design
12.2 Capacity Management Systems
12.3 Service Fulfillment
12.4 Design and Engineering
12.5 Summary
References
Part III: Transformation
Chapter 13: Integration and Innovation
13.1 Technology Integration
13.2 Lifecycle Support
13.3 Invention and Innovation
13.4 Summary
References
Chapter 14: Disasters and Outages
14.1 Disasters
14.2 Outages
14.3 The Vicious Cycle
14.4 Summary
References
Chapter 15: Technologies that Matter
15.1 Convergence or Conspiracy?
15.2 Technologies Beyond 2012
15.3 HTML5 and WEBRTC
15.4 Summary
References
Chapter 16: Carriers Transformed
16.1 Historical Transformations
16.2 Regulation and Investment
16.3 Consumer Wireline Networks and Services
16.4 Wireless Networks and Services
16.5 Backbone Networks
16.6 Science and Technology Matter
References
Appendix A: IPv6 Technologies
Appendix B: The Next Generation Network and Why We'll Never See It1
B.1 Claims of the Next Generation Network
B.2 Forces of Network Transformation
Acknowledgments
Index
This edition first published 2013
© 2013 John Wiley and Sons Ltd
Registered office
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom
For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com.
The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.
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 or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data
Cambron, G. Keith.
Global networks : engineering, operations and design / G. Keith Cambron.
pages cm
Includes bibliographical references and index.
ISBN 978-1-119-94340-2 (hardback)
1. Wireless communication systems. 2. Telecommunication. 3. Globalization. I. Title.
TK5103.C36 2012
384.5–dc23
2012022584
A catalogue record for this book is available from the British Library.
ISBN: 9781119943402
Dedicated to Amos E. Joel, Jr and the Members of the Technical Staff at AT&T Labs and SBC Labs
List of Figures
Figure 1.1 Simplex operation
Figure 1.2 Duplex model
Figure 1.3 Virtuous Cycle
Figure 1.4 Network and management systems
Figure 2.1 Telco system model
Figure 2.2 Full backplane chassis
Figure 2.3 Midplane chassis
Figure 2.4 T1 line card
Figure 2.5 System controller
Figure 2.6 ATCA chassis
Figure 2.7 Blade center chassis
Figure 2.8 Functional blade center diagram
Figure 3.1 Poorly behaved systems
Figure 3.2 Four port Ethernet NIC
Figure 3.3 Managed object hierarchy
Figure 3.4 Network element alarm processes
Figure 3.5 Link out of service example
Figure 3.6 Provisioning a new port
Figure 3.7 Data management hierarchy
Figure 4.1 Network impairment allocations
Figure 4.2 Enterprise VoIP network
Figure 5.1 Carrier functional model
Figure 5.2 Local access transport area
Figure 5.3 Local exchange SS7 network
Figure 5.4 Distribution area concept
Figure 5.5 SAI cross connect and serving terminal
Figure 5.6 Power spectral density for digital services
Figure 5.7 ADSL\emdash VDSL cable sharing
Figure 5.8 Bridge tap
Figure 5.9 ADSL aggregation network
Figure 5.10 ADSL2$+$ and VDSL aggregation network
Figure 5.11 Passive Optical Network (PON)
Figure 5.12 Single Family Unit (SFU) Optical Network Terminal (ONT)
Figure 5.13 Multiple Dwelling Unit (MDU) Optical Network Unit (ONU)
Figure 5.14 Pre-engineered optical distribution
Figure 5.15 FTTH aggregation network
Figure 5.16 HFC fiber node design
Figure 5.17 Mobile network architecture
Figure 5.18 GSM voice mobility management
Figure 5.19 SMS architecture
Figure 5.20 GPRS data mobility management
Figure 5.21 UMTS data mobility management
Figure 5.22 LTE and UMTS interworking
Figure 5.23 Adaptive modulation example
Figure 6.1 Backbone network regional view
Figure 6.2 Transport and routing
Figure 6.3 Architect's network view
Figure 6.4 Topological network view
Figure 6.5 SONET nodes
Figure 6.6 SONET ring
Figure 6.7 IP routing core
Figure 6.8 Core routing protocols
Figure 7.1 Carrier three layer model
Figure 7.2 Intelligent DNS
Figure 7.3 Intelligent Route Service Control Point (IRSCP)
Figure 7.4 IMS architecture
Figure 8.1 IXC call completion via LNP
Figure 8.2 Internet peering arrangements
Figure 8.3 Interdomain routing
Figure 8.4 SMS interworking architecture
Figure 9.1 Tiered operations support
Figure 9.2 AT\&T Global Network Operations Center (NOC)
Figure 10.1 Historical chart of consumer data rates
Figure 11.1 Network and support system roles
Figure 11.2 Hierarchy of management systems
Figure 11.3 Model NMS
Figure 11.4 TL1 management example
Figure 11.5 SNMP management example
Figure 11.6 CORBA example
Figure 11.7 Darkstar
Figure 11.8 Watch7 SS7 network monitor
Figure 11.9 Network management domains
Figure 11.10 Network management and the Virtuous Cycle
Figure 12.1 TMN functional layer model
Figure 12.2 TMN operations process model
Figure 12.3 Capacity management system design
Figure 12.4 High usage engineering
Figure 12.5 Demand and capacity chart
Figure 12.6 Service fulfillment
Figure 13.1 Fully developed test facility
Figure 14.1 SS7 networks circa 1991
Figure 14.2 DSC STP logical model
Figure 14.3 Traffic discard in a well-behaved system
Figure 14.4 The Vicious Cycle
About the Author
Keith Cambron has a broad range of knowledge in telecommunications networks, technology and design and R&D management. His experience ranges from circuit board and software design to the implementation of large public networks.
Keith served as the President and CEO of AT&T Labs, Inc. AT&T Labs designs AT&T's global IP, voice, mobile, and video networks. Network technology evaluation, certification, integration, and operational support are part of the Lab's responsibilities. During his tenure AT&T Labs had over 2000 employees, including 1400 engineers and scientists. Technologies produced by Labs ranged from core research to optical transport, IP routing, voice, and video systems.
2003 to 2006—Cambron served as the President and CEO of SBC Laboratories, Inc. The organization, which set the strategic technology objectives of SBC, was structured into four technology areas; Broadband, Wireless, Network Services, and Enterprise IT. SBC Labs led the industry in the introduction of VDSL and IPTV technologies.
1998 to 2003—Cambron was principal of Cambron Consulting, where he provided network and software design consulting services to the telecommunications industry. Working with clients such as SBC, Vodafone Wireless, Coastcom and various enterprise networks, Cambron designed and developed network management systems, a wireless Short Messaging Service (SMS) server, a Service Switching Point (SSP), and an ADSL transmission performance optimization system.
1987 to 1997—Cambron held leadership positions at Pacific Bell Broadband, acting as the chief architect of a network engineering team that developed a 750 MHz hybrid fiber/coax-based network. For this project, Cambron received Telephony's “Fiber in the Loop” design award.
His career started at Bell Telephone Laboratories in 1977, where he began as a member of the technical staff. He advanced to Director of Local Switching Systems Engineering and led a team to design automated verification test tools for local digital switch testing. Cambron went on to become Director of Network Systems Verification Testing at Bell Communications Research, heading field verification teams in all seven Regional Bell Operating Companies to test “first in nation” technologies, including the first local digital switching systems.
Cambron has been profiled in Telephony and America's Network, and has published in IEEE Communications and Proceedings of the IEEE. He taught Object Oriented Design at Golden Gate University in San Francisco and is a Senior Member of the Institute of Electrical and Electronics Engineers (IEEE).
In 2010, Cambron was named by CRN Magazine as one of the Top 25 Technology Thought-Leaders in the world. Keith received IEEE Communications Society's Chairman's Communication Quality and Reliability Award in 2007. He holds ten patents for telecommunications software and systems he designed and deployed.
Cambron received his BS in Electrical Engineering from the University of Missouri, an MS in Systems Management from the University of Southern California, and a Programming Certification from the University of California at Berkeley. He is a retired Commander in the United States Naval Reserve.
Foreword
Networks today are like the air we breathe, so ubiquitous we often take them for granted and in fact don't even realize they're there. Whether we are working, studying, communicating or being entertained, we rely on networks to make whatever we need to happen, happen. This trend is increasing as networks become more and more powerful and reach more deeply into the fabric of our lives.
This reach is not limited to just the wealthy or to developed nations, however, as lower costs and higher capacity extend the power of networks to citizens all around the globe. That's what makes this book so relevant and so timely. A clear understanding of these networks is essential for those that would design, construct, operate and maintain them. As Keith points out early in this volume, the growing gap between the academic description of networks and the real world design and operation of these networks, is a key divide that needs bridging. And Keith is in a unique position to do this.
I've known Keith for over 15 years, and have always found him to be a fascinating and indeed remarkable man. His curiosity and intelligence, coupled with a career so deeply involved in network design at AT&T has given him the tremendous insight that he shares in this book. Keith has never been afraid to step outside the accepted norm, if he felt the need, for pursuit of a new area of excellence. This is what makes his knowledge and understanding so valuable and drives the core of this work.
Looking forward, Moore's Law will continue to enable the exponential growth of the value of the underlying technologies, namely processing, memory and optical communications speed, that make these networks tick. The resultant capabilities of the next generations of networks, five years or a decade out, are virtually indescribable today! That in the end is what makes this book so valuable—a thorough understanding of the design principles described herein will allow those that shape our networks in the future to “get it right,” enhancing our lives in ways we cannot begin to imagine.
Robert C. McIntyreChief Technical Officer, Service Provider GroupCisco Systems
Preface
When I began my career in telecommunications in 1977 at Bell Telephone Laboratories two texts were required reading, Engineering and Operations in the Bell System [1] and Principles of Engineering Economics [2]. Members of Technical Staff (MTS) had Masters or PhDs in engineering or science but needed grounding in how large networks were designed and operated, and how to choose designs that were not only technically sound, but economically viable. As the designers of the equipment, systems and networks, engineers at Bell Labs were at the front end of a vertically integrated company that operated the US voice and data networks. Operational and high availability design were well developed disciplines within Bell Labs, and network systems designs were scrutinized and evaluated by engineers practicing in those fields. So ingrained in the culture was the operational perspective, that engineers and scientists were strongly encouraged to rotate through the Operating Company Assignment Program (OCAP) within the first two years of employment. During that eight week program engineers left their Bell Labs jobs and rotated through departments in a Bell Telephone Operating Company, serving as operators, switchmen, installers and equipment engineers. OCAP was not restricted to engineers working on network equipment; members of Bell Labs Research participated in the program. AT&T was not alone in recognizing the value of operational and reliability analysis in a vertically integrated public telephone company, Nippon Telephone and Telegraph, British Telecom, France Telecom and other public telephone companies joined together in technical forums and standards organizations to codify operational and high availability design practices.
After 1982 regulatory, technology and market forces dramatically changed the way networks and systems were designed and deployed. Gone are vertically integrated franchise operators, replaced by interconnected and competing networks of carriers, equipment and systems suppliers, and integrators. Innovation, competition and applications are the engines of change; carriers and system suppliers respond to meet the service and traffic demands of global networks growing at double and even triple digit rates, carrying far more video content than voice traffic. Consumer and enterprise customers are quick to adopt new devices, applications and network services; however, when legacy carriers deliver the service the customers' expectations for quality and reliability are based on their long experience with the voice network. The industry has largely delivered on those expectations because an experienced cadre of engineers from Bell Labs and other carrier laboratories joined startups and their spun off suppliers like Lucent and Nortel. But as time passes, the operational skill reservoir recedes not only because the engineers are retiring, but because of the growing separation between engineers that design and operate networks, and those that design equipment, systems and applications that enter the network. The clearest example of the change is the influx of IT trained software engineers into the fields of network applications and systems design. Experience in the design of stateless web applications or financial systems are insufficient for the non-stop communication systems in the network that continue to operate under a range of hardware faults, software faults and traffic congestion.
My own journey gave me a front row seat through the transformation from a regulated voice network to a competitive global IP network. As a systems engineer in the 1970s I worked on call processing requirements and design for the No. 1 ESS. In the 1980s I led teams of test and verification engineers in the certification of the DMS-10, DMS-100, No. 5ESS, No. 2 STP, DMS STP and Telcordia SCP. I also led design teams building integrated test systems for Signaling System No. 7 and worked for startup companies designing a DS0/1/3 cross connect system and a special purpose tandem switching system. During the last eight years I headed SBC Labs and then AT&T Labs as President and CEO. Working with engineers across network and systems, and spending time with faculty and students at universities I became aware of the growing gap in operational design skills. Universities acknowledge and reward students and faculty for research into theoretical arenas of network optimization and algorithm design. Their research is seldom based on empirical data gathered from networks, and rarer still is the paper that actually changes the way network or systems operate. I chose to write this book to try and fill some of that void. My goal is to help:
those students and faculty interested in understanding how operational design practices can improve system and network design, and how networks are actually designed, managed and operated;
hardware and software engineers designing network and support systems;
systems engineers developing requirements for, or certifying network equipment;
systems and integration engineers working to build or certify interfaces between network elements and systems;
operations support systems developers designing software for the management of network systems; and
managers working to advance the skills of their engineering and operating teams.
The book is organized into three parts; Networks, Teams and Systems, and Transformation. It is descriptive, not prescriptive; the goal is not to tell engineers how to design networks but rather describe how they are designed, engineered and operated; the emphasis is on engineering and design practices that support the work groups that have to install, engineer and run the networks. Areas that are not addressed in the book are network optimization, engineering economics, regulatory compliance and security. Security as a service is described in the chapter on cloud services but there are several texts that better describe the threats to networks and strategies for defense [3–4].
References
1. Engineering and Operations in the Bell System, 1st edn, AT&T Bell Laboratories (1977).
2. Grant, E.L. and Ireson, W.G. (1960) Principles of Engineering Economy, Ronald Press Co., New York.
3. Cheswick, W.R. Bellovin, S.M. and Rubin, A.D. (2003) Firewalls and Internet Security, 2nd edn, Repelling the Wily Hacker, Addison Wesley.
4. Amoroso, E. (2010) Cyber Attacks: Protecting National Infrastructure, Butterworth-Heinemann, November.
Acknowledgments
The technical breadth of this text could not have been spanned without the help of engineers I have had the privilege of working with over the years. While I researched and wrote the entire text, these contributors were kind enough to review my material. I am grateful for the contributions of John Erickson, Mike Pepe, Chuck Kalmanek, Anthony Longhitano, Raj Savoor, and Irene Shannon. Each reviewed specific chapters that cover technology within their area of expertise and corrected my technical errors and omissions. They are not responsible for the opinions and projections of future technology trends, and any remaining errors are mine.
I also want to thank the team at John Wiley & Sons, Ltd for guiding me through the writing and publishing process. They made the experience enjoyable and their professional guidance kept me on a sure track.
List of Acronyms
10G
10 Gigabit
100G
100 Gigabit
10GEPON
10 Gigabit Ethernet Passive Optical Network
21CN
21st Century Network
3G
Third Generation Mobile Technology
3GPP
Third Generation Partnership Project
4G
Fourth Generation Mobile Technology
40G
40 Gigabit
400G
400 Gigabit
6rd
IPv6 Rapid Deployment
AAAA
Quad A DNS Record
ABR
Available Bit Rate
ACD
Automatic Call Distributor
ACM
Address Complete Message (ISUP)
ADL
Advanced Development Lab
ADM
Add-Drop Multiplexer
ADPCM
Adaptive Differential Pulse Code Modulation
ADSL
Asymmetric Digital Subscriber Line
ADSL1
Asymmetric Digital Subscriber Line G.992.1 standard
ADSL2+
Asymmetric Digital Subscriber Line G.992.5 standard
AIN
Advanced Intelligent Network
AINS
Automatic In-Service
AIS
Alarm Indication Signal
ALG
Application Level Gateway
ALI
Automatic Line Identification
AMI
Alternate Mark Inversion
AMPS
Advanced Mobile Phone Service
AMR
Adaptive Multi-Rate
API
Application Programming Interface
APN
Access Point Name
APS
Automatic Protection Switching
ARGN
Another Really Good Network
ARP
Address Resolution Protocol
AS
Autonomous System
AS
Application Server
ASON
Automatic Switched Optical Network
ASP
Application Service Provider
AT
Access Tandem
ATA
Analog Terminal Adapter
ATCA
Advanced Telecommunications Computing Architecture
ATM
Asynchronous Transfer Mode
ATSC
Advanced Television Systems Committee
AUMA
Automatic and Manual Service State
AWG
American Wire Gauge
AWS
Advanced Wireless Services
BCP
Business Continuity Plan
BCPL
Basic Combined Programming Language
BGCF
Breakout Gateway Control Function
BGF
Border Gateway Function
BGP
Border Gateway Protocol
BITS
Building Integrated Timing Supply
BLSR
Bi-directional Line Switched Ring
BORSCHT
Battery, Over-voltage, Ringing, Supervision, Codec, Hybrid, Testing
BPON
Broadband Passive Optical Network
BRAS
Broadband Remote Access Server
BRI
Basic Rate Interface (ISDN)
BSC
Base Station Controller
BSD
Berkeley Software Distribution
BSS
Business Support System
BSSMAP
Base Station Subsystem Mobile Application Part
BT
British Telecom
BTL
Bell Telephone Laboratories
BTS
Base Transceiver Station
CALEA
Communications Assistance for Law Enforcement Act
CAMEL
Customized Applications for Mobile network Enhanced Logic
CAS
Channel Associated Signaling
CAT3
Category 3, refers to a grade of twisted pair cable
CATV
Community Antenna Television
CBR
Constant Bit Rate
CCAP
Converged Cable Access Platform
CCIS
Common Channel Interoffice Signaling
CCS
Common Channel Signaling
CDB
Centralized Database
CDF
Charging Data Function
CDMA
Code Division Multiple Access
CDN
Content Delivery Network
CDR
Call Detail Record
CE
Customer Edge
CES
Circuit Emulation Service
CGN
Carrier Grade NAT
CGN64
Carrier Grade NAT IPv6/IPv4
CIC
Carrier Identification Code
CIC
Circuit Identification Code
CLASS
Custom Local Area Signaling Services
CLEC
Competitive Local Exchange Carrier
CLI
Command Line Interface
CLLI
Common Language Location Identifier
CM
Cable Modem
CM
Capacity Management
CMS
Customer Management System
CMTS
Cable Modem Termination System
CNAM
Calling Name Service
CO
Central Office
CONF
Conference Services
CORBA
Common Object Request Broker Architecture
CoS
Class of Service
CPE
Customer Premises Equipment
CPU
Central Processing Unit
CR
Constrained Routing
CRC
Cyclic Redundancy Check
CRS
Carrier Routing System
CSCF
Call Session Control Function
CSFB
Circuit Switched Fallback
CSS3
Cascading Style Sheet 3
CTAG
Command Tag
CURNMR
Current Noise Margin
DA
Distribution Area
DAML
Digitally Added Main Line
DARPA
Defense Advanced Research Projects Agency
DAS
Directed Antenna System
DBMS
Database Management System
DBOR
Database of Record
DCC
Data Communications Channel
DCS
Digital Cross Connection System
DHCP
Dynamic Host Control Protocol
DHCP6
Dynamic Host Control Protocol for IPv6
DLC
Digital Loop Carrier
DLNA
Digital Living Network Alliance
DMS
Digital Multiplex System
DMT
Discrete Multitone
DMTS
Distinguished Member of Technical Staff
DNS
Domain Name System
DNS64
Domain Name System for IPv4 and IPv6
DOCSIS
Data Over Cable Service Interface Specification
DoS
Denial Of Service
DPM
Defects Per Million
DSBLD
Disabled Service State
DSL
Digital Subscriber Line
DSLAM
Digital Subscriber Line Access Multiplexer
DSM
Dynamic Spectrum Management
DSP
Digital Signal Processor
DSTM
Dual Stack IPv6 Transition Mechanism
DSX
Digital Cross Connect
DTAP
Direct Transfer Application Part (SS7)
DTV
Digital Television
DVB
Digital Video Broadcast
DVD
Digital Video Disc
DVR
Digital Video Recorder
DWDM
Dense Wave Division Multiplexing
E911
Enhanced 911
EADAS
Engineering Admin Data Acquisition System
EDFA
Erbium Doped Fiber Amplifier
EDGE
Enhanced Data Rates for Global Evolution
EFM
Ethernet in the First Mile
EGP
External Gateway Protocol
EIGRP
Enhanced Interior Gateway Routing Protocol
EMEA
Europe, the Middle East and Africa
EMS
Element Management System
ENUM
E.164 Number Mapping
EOC
Embedded Operations Channel
EPON
Ethernet Passive Optical Network
ESAC
Electronic Systems Assurance Center
ESME
External Short Messaging Entity
ESS
Electronic Switching System
eTOM
Enhanced Telecom Operations Map
ETS
Electronic Translator System
FCAPS
Fault, Configuration, Accounting, Performance, Security
FCC
Federal Communications Commission
FDD
Frequency Division Duplex
FDMA
Frequency Division Multiple Access
FEC
Forwarding Equivalent Class
FEXT
Far End Crosstalk
FOU
Field of Use
FRR
Fast Reroute
FRU
Field Replaceable Unit
FSAN
Full Service Access Network
FTP
File Transfer Protocol
FTTB
Fiber To The Building
FTTC
Fiber To The Curb
FTTH
Fiber To The Home
FTTN
Fiber To The Node
GEM
GPON Encapsulation Method
GERAN
GSM EDGE Radio Access Network
GGSN
Gateway General Packet Radio Services Support Node
GMPLS
Generalized Multi-protocol Label Switching
GMSC
Gateway Mobile Switching Center
GMSK
Gaussian Minimum Shift Keying
GNOC
Global Network Operations Center
GPON
Gigabit Passive Optical Network
GPS
Global Positioning System
GRE
Generic Routing Encapsulation
GRX
GPRS Routing Exchange
GSM
Global System for Mobile Communications
GTP
GPRS Tunneling Protocol
GTT
Global Title Translation
HD
High Definition
HDSL
High Bitrate Digital Subscriber Line
HDTV
High Definition Television
HFC
Hybrid Fiber Coax
HLR
Home Location Register
HPNA
Home Phone line Networking Alliance
HR
Human Resources
HSDPA
High Speed Downlink Packet Access
HSPA
High Speed Packet Access
HSS
Home Subscriber Server
HSUPA
High Speed Uplink Packet Access
HTML
Hyper Text Markup Language
HTTP
Hyper Text Transfer Protocol
HVAC
Heating, Ventilation and Air Conditioning
IAM
Initial Address Message (SS7)
IAS
Internet Access Service
IBCF
Interconnection Border Control Function
ICMP
Internet Control Message Protocol
IDL
Interface Definition Language
IGMP
Internet Group Management Protocol
IGP
Interior Gateway Protocol
ILEC
Incumbent Local Exchange Carrier
IM
Instant Messaging
IMS
IP Multimedia Subsystem
IMSI
International Mobile Subscriber Identifier
IN
Intelligent Network
IOT
Interoperability Testing
IP
Internet Protocol
IPMI
Intelligent Platform Management Interface
IPTV
Internet Protocol Television
IPX
Internet Protocol Packet Exchange
IRAT
Inter-Radio Access Technology
IRSCP
Intelligent Route Service Control Point
IS
In-Service
ISATAP
Intra-Site Automatic Tunnel Addressing Protocol
ISDN
Integrated Services Digital Network
ISP
Internet Services Provider
ISUP
ISDN User Part
IT
Information Technology
ITP
IP Transfer Point
IVR
Interactive Voice Response
IXC
Interexchange Carrier
IXP
Internet Exchange Point
KPI
Key Performance Indicator
LAN
Local Area Network
LATA
Local Access Transport Area
LCP
Local Convergence Point
LD
Long Distance
LDP
Label Distribution Protocol
LEC
Local Exchange Carrier
LEN
Line Equipment Number
LER
Label Edge Router
LERG
Local Exchange Routing Guide
LFIB
Label Forwarding Information Base
LFO
Line Field Organization
LIDB
Line Information Database
LLDP
Local Loop Demarcation Point
LMTS
Lead Member of Technical Staff
LNP
Local Number Portability
LOF
Loss of Frame
LOL
Loss of Link
LOS
Loss of Signal
LP
Link Processor
LPBK
Loop Back
LPR
Loss of Power
LRF
Location Retrieval Function
LRN
Local Routing Number
LSA
Link State Advertisement
LSDB
Link State Database
LSN
Large Scale NAT
LSP
Label Switched Path
LSR
Label Switch Router
LSSGR
LATA Switching System Generic Requirements
LTE
Long Term Evolution
MA
Manual Service State
MAP
Mobile Application Part
MDF
Main Distribution Frame
MDR
Message Detail Record
MDU
Multiple Dwelling Unit
MED
Multi-Exit Discriminator
MF
Multi-Frequency
MFJ
Modified Final Judgment
MGCF
Media Gateway Control Function
MGW
Media Gateway
MIB
Management Information Base
MIME
Multipurpose Internet Mail Extension
MIMO
Multiple In Multiple Out
MOB
Mobility and Location Services
MME
Mobile Management Entity
MML
Man Machine Language
MMS
Multimedia Message Service
MNO
Mobile Network Operator
MOP
Method of Procedure
MPEG
Motion Pictures Expert Group
MPLS
Multiprotocol Label Switching
MPOE
Minimum Point of Entry
MRFC
Media Resource Function Controller
MRFP
Media Resource Function Processor
MS
Mobile Station
MSC
Mobile Switching Center
MSIN
Mobile Subscriber Identification Number
MSISDN
Mobile Subscriber Integrated Services Digital Subscriber Number
MSO
Multiple System Operator
MSPP
Multiservice Provisioning Platform
MSR
Multi-standard Radio
MSRN
Mobile Station Routing Number
MT
Maintenance Service State
MTS
Member of Technical Staff
MTSO
Mobile Telephone Switching Office
NAP
Network Access Point
NAT
Network Address Translation
NB
Narrowband
NCL
Network Certification Lab
NCP
Network Control Point
NDC
Network Data Center
NE
Network Element
NEBS
Network Equipment Building Standards
NEXT
Near End Crosstalk
NGN
Next Generation Network
NIC
Network Interface Card
NICE
Network-Wide Information Correlation and Exploration
NID
Network Interface Device
NLRI
Network Layer Reachability Information
NMC
Network Management Center
NMP
Network Management Plan
NMS
Network Management System
NNI
Network to Network Interface
NOC
Network Operations Center
NORS
Network Outage Reporting System
NPA
Numbering Plan Area
NPOE
Network Point of Entry
NPRM
Notice of Proposed Rule Making
NR
Normal Service State
NSE
Network Systems Engineering
NSTS
Network Services Test System
NTSC
National Television Systems Committee
NTT
Nippon Telephone and Telegraph
OA&M
Operations, Administration & Maintenance
OEM
Original Equipment Manufacturer
OFDM
Orthogonal Frequency Division Multiplexing
OFDMA
Orthogonal Frequency Division Multiple Access
OID
Object Identifier
OLT
Optical Line Terminal
OMA
Open Mobile Alliance
ONT
Optical Network Terminal
ONU
Optical Network Unit
OOS
Out of Service
ORB
Object Request Broker
ORT
Operational Readiness Test
OS
Operating System
OSA
Open Services Architecture
OSI
Open Systems Interconnection
OSP
Outside Plant
OSPF
Open Shortest Path First
OSS
Operations Support System
OTA
Over the Air
OTDR
Optical Time Domain Reflectometer
OTN
Optical Transport Network
OTT
Over The Top
PAL
Phase Alternating Line video standard
PAT
Port Address Translation
PBX
Private Branch Exchange
PC
Personal Computer
PCEF
Policy Charging Enforcement Function
PCM
Pulse Code Modulation
PCRF
Policy Charging Rules Function
PCS
Personal Communication Service
PCU
Packet Control Unit
PDN
Packet Data Network
PDP
Packet Data Protocol
PDU
Packet Data Unit
PE
Provider Edge
PEG
Public Education and Government
PERF
Policy Enforcement Rules Function
PIC
Polyethylene Insulated Cable
PIC
Primary Inter-LATA Carrier
PIM
Protocol Independent Multicast
PLMN
Public Land Mobile Network
PM
Performance Management
PMTS
Principal Member of Technical Staff
PON
Passive Optical Network
POP
Point of Presence
POTS
Plain Old Telephone Service
PPP
Point to Point Protocol
PRI
Primary Rate Interface (ISDN)
PSAP
Public Service Answering Point
PSL
Production Support Lab
PSTN
Public Switched Telephone Network
PTE
Path Terminating Equipment
PUC
Public Utility Commission
PVC
Private Virtual Circuit
RAB
Radio Access Bearer
RADIUS
Remote Authentication Dial In User Service
RAN
Radio Access Network
RBOC
Regional Bell Operating Company
RCA
Root Cause Analysis
RDC
Regional Data Center
RF
Radio Frequency
RFC
Request for Comment
RFP
Request for Proposal
RIR
Regional Internet Registry
RNC
Radio Network Controller
ROADM
Reconfigurable Optical Add Drop Multiplexer
RP
Route Processor
RRC
Radio Resource Control
RSS
Remote Switching System
RSVP
Resource Reservation Protocol
RTM
Rear Transition Module
RTP
Real-time Transport Protocol
RTT
Round Trip Time
SAI
Serving Area Interface
SAN
Storage Area Network
SBC
Session Border Controller
SCC
Switching Control Center
SCCS
Switching Control Center System
SCE
CAMEL Service Environment
SCP
Service Control Point
SCTE
Society of Cable and Television Engineers
SDH
Synchronous Digital Hierarchy
SDSL
Symmetric Digital Subscriber Line
SDV
Switched Digital Video
SECAM
Sequential Color with Memory (FR)
SEG
Security Gateway
SELT
Single Ended Line Test
SFTP
Secure File Transfer Protocol
SFU
Single Family Unit
SGSN
Serving General Packet Radio Services Node
SGW
Signaling Gateway
SIL
Systems Integration Lab
SIM
Subscriber Identity Module
SIP
Session Initiation Protocol
SLA
Service Level Agreement
SLF
Subscription Locator Function
SMI
SNMP Structure of Management Information
SMIL
Synchronized Multimedia Integration Language
SMPP
Short Messaging Peer to Peer Protocol
SMPTE
Society of Motion Picture and Television Engineers
SMS
Short Message Service
SMSC
Short Message Service Center
SMTP
Simple Mail Transfer Protocol
SNMP
Simple Network Management Protocol
SNR
Signal to Noise Ratio
SOA
Service Oriented Architecture
SOAP
Simple Object Access Protocol
SONET
Synchronous Optical Network
SP
Signaling Point
SPC
Stored Program Control
SPF
Shortest Path First
SPOI
Signalng Point of Interface
SQL
Standard Query Langauge
SS6
Signaling System No. 6
SS7
Signaling System No. 7
SSF
Service Switching Function
SSH
Secure Shell
SSP
Service Switching Point
STB
Settop Box
STE
Section Terminating Equipment
STM
Synchronous Transport Module
STP
Signaling Transfer Point
STS
Synchronous Transport Signal
SUT
System Under Test
TAC
Technical Assistance Center
TAS
Telephony Application Server
TCAP
Transaction Capabilities Part
TCP
Transmission Control Protocol
TDM
Time Division Multiplex
TDMA
Time Division Multiple Access
TE
Traffic Engineering
TID
Target Identifier
TL1
Transaction Language 1
TMF
Telecommunications Management Forum
TMN
Telecommunications Management Network
TNMR
Target Noise Margin
TOD
Time of Day
TRAU
Transcoding and Rate Adaption Unit
TrGW
Transition Gateway
TSD
Technical Service Description
TSI
Time Slot Interchange
TTL
Time to Live
UAS
Unassigned service state
UAT
User Acceptance Test
UBR
Undefined Bit Rate
UDP
User Datagram Protocol
UE
User Equipment
UEQ
Unequipped service state
UHDTV
Ultra-High Definition TV
ULH
Ultra Long Haul
UML
Uniform Modeling Language
UMTS
Universal Mobile Telecommunications System
UNE
Unbundled Network Element
UNI
User to Network Interface
UPSR
Unidirectional Path Switched Rings
URI
Uniform Resource Identifier
URL
Uniform Resource Locator
UTC
Universal Coordinated Time
UTRAN
UMTS Radio Access Network
VBR
Variable Bit Rate
VC
Virtual Circuit
VDSL
Very High Bit Rate Digital Subscriber Line
VHO
Video Home Office
VLAN
Virtual Local Area Network
VLR
Visiting Local Register
VOD
Video On Demand
VP
Virtual Path
VPLS
Virtual Private LAN Service
VPN
Virtual Private Network
VRF
Virtual Routing and Forwarding
WAN
Wide Area Network
WAP
Wireless Application Protocol
WB
Wideband
WLAN
Wireless Local Area Network
WSDL
Web Services Description Language
XML
Extensible Markup Language
YAMS
Yet Another Management System
Part I
Networks
Chapter 1
Carrier Networks
We have come a long way in a short time. Instant communication arrived relatively recently in the history of man, with the invention of the telegraph at the beginning of the nineteenth century. It took three quarters of a century before we saw the first major improvement in mass communication, with the arrival of the telephone in 1874, and another half century before a national network and transcontinental communication became common in the US. But what was a middle class convenience years ago is now a common necessity. Today, our worldwide communication Network is a model of egalitarian success, reaching into the enclaves of the wealthy and the street vendors in villages of the developing world with equal ease. Remarkably it remains largely in the hands of the private sector, and is held together and prospers through forces of cooperation and competition that have served society well.
The Network is made up of literally millions of nodes and billions of connections, yet when we choose to make a call across continents or browse a web site in cyberspace we just expect it to work. I use the proper noun Network when referring to the global highway of all interconnection carrier networks, such as Nippon Telephone and Telegraph (NTT), British Telecom (BT), China Telecom, AT&T, Verizon, Deutsche Telekom, Orange, Hurricane Electric, and many others, just as we use the proper noun Internet to refer to the global public Internet Protocol (IP) network. The Internet rides on the Network. If the Network were to fail, even within a city, that city would come to a halt. Instead of purchasing fuel at the pump with a credit card, drivers would line up at the register while the attendant tried to remember how to make change instead of waiting for the Network to verify a credit card. Large discount retail outlets would have their rows of registers stop and for all practical purposes the retailers would close their doors. Alarm systems and 911 emergency services would cease to function. Streets would become congested because traffic lights would no longer be synchronized. So what are the mechanisms that keep the Network functioning 24 hours a day with virtually no failures?
1.1 Operating Global Networks
The global nature of networks is a seismic change in the history of modern man. In the regulated world of the past franchise carriers completely dominated their national networks. Interconnection among networks existed for decades, but carriers did not over build each other in franchise areas. That all changed in the latter decades of the twentieth century as regulation encouraged competition and data services emerged. International commerce and the rise of multinational companies created a demand for global networks operated by individual carriers. Multinational companies wanted a single operator to be held accountable for service worldwide. Many of them simply did not want to be in the global communications business and wanted a global carrier to sort through interconnection and operations issues inherent in far reaching networks.
In parallel with globalization was the move to the IP. The lower layers of the Open System Interconnection (OSI) protocol stack grew because of global scale, and upper layer complexity; the complexity increased with new services such as mobility, video, and the electronic market, largely spurred by Internet services and technology. Operators were forced to reexamine engineering and operating models to meet global growth and expanding service demand. Before deregulation reliability and predictability were achieved through international standards organizations, large operating forces, and highly structured and process centric management regimes. Deregulation, competition, global growth, and service expansion meant that model was no longer economic and could not respond to the rapid introduction of new services and dramatic growths in traffic.
Operating models changed by applying the very advances in technology which drove demand. Reliable networks were realized by reducing the number of failures, by shortening the time for repair, or both. In the old model central offices were staffed with technicians that could respond on short notice to failures, keeping restoral times low. In the new model networks are highly redundant, well instrumented, constantly monitored, and serviced by a mobile work force.
1.1.1 The Power of Redundancy
This section introduces the foundation of global network reliability, redundancy using a simple systems model.
1.1.1.1 Simplex Systems
In the model following, a subscriber at A sends information i0 to a subscriber at B. The information arrives at B via a communications system S0 as i1 after a delay of t (see Figure 1.1).
Figure 1.1 Simplex operation.
Subscribers care about two things, the fidelity of the information transfer and transmission time. Fidelity means that the information received, i1, should be as indistinguishable from the information sent, i0, as possible. If we assume for simplicity that our communication depends on a single system, S0, that fails on average once every year, and it takes 4 h to get a technician on site and restore service, the service will be down for 4 h each year on average, yielding a probability of failure of 4.6 × 10−5, or an availability of 99.954%. That means we expect to fail about once for every 2000 attempts. For most communications services that is a satisfactory success rate.
But real world connections are composed of a string of systems working in line, possibly in the hundreds, any one of which can fail and impede the transfer. For a linear connection of 100 such systems, our failure probability grows to 4.5 × 10−3 and availability drops to 95.5%. Approximately 1 in 20 attempts to use the system will fail.
1.1.1.2 Redundant Systems
The chances of success can be dramatically improved by using a redundant or duplex system design, shown in Figure 1.2.
Figure 1.2 Duplex model.
In the design two identical systems, S0 and S1 are each capable of performing the transfer. One is active and the other is on standby. Since only one system affects the transfer, some communication is needed between the systems and a higher authority is needed to decide which path is taken.
In the duplex system design the probability of failure drops to 2.1 × 10−5 for 100 systems in line, an improvement of more than 100× for an investment of 2×. Availability rises to 99.998%. We expect to fail only once in each 50 000 attempts.
Implicit in the model are some key assumptions.
Failures are random and non-correlated. That is the probability of a failure in S
1
is unrelated to any failure experienced by S
0
. Since it's likely the designs of the two systems are identical, that assumption may be suspect.
The intelligence needed to switch reliably and timely between the two systems is fail-safe.
When S
0
fails, Operations will recognize it and take action to repair the system within our 4 h timeframe.
1.1.1.3 Redundant Networks
Redundancy works within network systems; their designs have two of everything essential to system health: power supplies, processors, memory, and network fabric. Adopting reliable network systems doesn't necessarily mean networks are reliable. Network systems have to be connected with each other over geographical expanses bridged by physical facilities to build serviceable networks. Physical facilities, optical fiber, telephone cable, and coaxial cable are exposed to the mischiefs of man and of nature. Dual geographically diverse routes to identical network systems preserve service if the end nodes recognize that one route has failed and the other is viable. Global networks rely on redundant systems within redundant networks. The combination is resilient and robust, providing any failure is recognized early and maintenance is timely and thorough.
The next sections explore this foundational model in more depth in an attempt to understand how it works, and how it can break down in real networks.
1.1.2 The Virtuous Cycle
In the 1956 film Forbidden Planet, an advanced civilization called the Krell invents a factory that maintains and repairs itself automatically. In the movie, although the Krell are long extinct, the factory lives on, perpetually restoring and repairing itself. Some academics and equipment suppliers promote this idea today using the moniker “self-healing network.” An Internet search with that exact phrase yields 96 000 entries in the result; it is a popular idea indeed. Academic papers stress mathematics, graphs, and simulations in search of elegant proofs of the concept. Yet real networks that perform at the top of their class do so because of the way people design, operate, and manage the technology. It is the blend of systems, operations, and engineering that determine success or failure. Systems and people make the difference. Figure 1.3 illustrates the Virtuous Cycle of equipment failure, identification, and restoral.
Figure 1.3 Virtuous Cycle.
The cycle begins at the top, or 12 o'clock, where the Network is operating in duplex that is full redundancy with primary and alternate paths and processes. Moving in a clockwise direction, a failure occurs signified by the X, and the Network moves from duplex to simplex operation, although no traffic is affected. While the Network is operating in simplex it is vulnerable to a second failure. But before operators can fix the problem they need to recognize it. Notification is the process whereby network elements send alarm notifications to surveillance systems that alert network operators to the situation. Notifications seldom contain sufficient information to resolve the problem, and in many situations multiple notifications are generated from a single fault. Operators must sort out the relevant notifications and sectionalize the fault to a specific network element in a specific location. The failed element can then be put in a test status, enabling operators to run diagnostics and find the root cause of the failure. Hardware faults are mitigated by replacing failed circuit packs. Software faults may require a change of configuration or parameters, or restarting processes. Systems can then be tested and operation verified before the system is restored to service, and the network returns to duplex operation. Later chapters explore these steps in detail.
1.1.3 Measurement and Accountability
The Virtuous Cycle enables highly trained people to work with network systems to restore complex networks quickly and reliably when they fail. But it does nothing to insure the network has sufficient capacity to handle demands placed upon it. It does not by itself give us any assurance the service the customer is receiving meets a reasonable standard. We can't even be sure network technicians are following the Virtuous Cycle diligently and restoring networks promptly. To meet these goals a broader system of measurements and accountability are needed. Carrier networks are only as good as the measurement systems and the direct line of measurements to accountable individuals. This is not true in smaller networks; when I speak with Information Technology (IT) and network engineers in smaller organizations they view carriers as having unwarranted overhead, rules, and systems. In small networks a few individuals have many roles and are in contact with the network daily. They see relationships and causality among systems quickly; they recognize bottlenecks and errors because of their daily touches on the network systems. Such a model does not scale. Carrier networks have hundreds of types of systems and tens of thousands of network elements. Carrier networks are more akin to Henry Ford's production lines than they are to Orville's and Wilbur's bicycle shop. Quality and reliability are achieved by scaling measurement and accountability in the following ways.
End service objectives
—identify measurable properties of each service; commit to service standards, communicate them, and put them into practice.
Network systems measurement
—using service objectives analyze each network and network element and set measurable objectives that are consistent with the end to end service standard.
Assign work group responsibility
—identify which work group is responsible for meeting each of the objectives and work with them to understand how they are organized, what skills they have and what groups they depend upon and communicate with regularly.
Design engineering and management systems
—systems should support people, not the other way round. Find out what systems the teams already use and build on those if at all possible. Don't grow YAMS (yet another management system).
1.2 Engineering Global Networks
Changes in operations as dramatic as they have been are greater yet for the design and engineering of global networks. Carriers in the US prior to 1982 were part of a vertically integrated company, AT&T® or as it was commonly known, the Bell System. AT&T General Departments operated the complete supply and operations chain for the US telecommunications industry. Wholly owned subsidiaries planned, designed, manufactured, installed, and operated the network. AT&T's integrated business included the operations support, billing, and business systems as well. Carriers (the operating companies) had no responsibility for equipment design or selection, and limited responsibility for network design. Today carriers have full responsibility for planning, designing, installing, and operating their networks. They also have a direct hand in determining the functionality and high level design of network systems, and operations and business systems. The sections that follow summarize responsibilities of carrier engineering departments.
1.2.1 Architecture
High level technology choices are the responsibility of Engineering. Engineering architects analyze competing technologies, topologies, and functional delegation to determine the merits and high level cost of each. Standards organizations such as ITU, IETF, and IEEE are forums serving to advance ideas and alternatives. Suppliers naturally promote new ideas and initiatives as well, but from their point of view. Long range plans generally describe the evolution of networks but may not address practical and necessary design and operational transition issues.
1.2.2 Systems Engineering
A wide range of responsibilities rest with systems engineers. They begin with high level architectural plans and translate them into detailed specifications for networks and for the individual network elements. Equipment recommendations, testing, certification, and integration are all performed by these engineers. Operational support, IT integration, and network design are performed by systems engineers as well.
1.2.3 Capacity Management
There are four general ways in which network capacity is expanded. Each is described in the following.
1.2.3.1 Infrastructure Projects
Periodically major network augmentation is undertaken for a variety of reasons.
Expansion into a new geography is a common trigger. A country adopts competitive rules that enable over building the incumbent.
Technology obsolescence, such as the shift from Frame Relay to IP networks leads to a phased introduction of new technology. The new networks often must interwork with the legacy technology making the transition more challenging.
Carrier mergers or acquisitions are followed by network rationalization and integration. The numbers and types of network elements are winnowed out to ease operational demands.
New lines of business, such as Internet Protocol Television (IPTV) or content distribution, place new demands on the network requiring specialized technology design and deployment.
1.2.3.2 Customer Wins
Major customer contract wins significantly increase demand at large customer locations, rendering the existing capacity inadequate. Sometimes outsourcing of a Fortune 500 company network can be accompanied by an agreement to transfer their entire network, employees, and systems to the winning carrier. If they are of sufficient scope, the accompanying network augmentations are treated as separate projects with dedicated engineering, operations, and finance teams.
1.2.3.3 Capacity Augmentation
By far the most common reason for adding equipment and facilities to a network is the continuous growth in demand of existing services and transport. For decades voice traffic grew at a rate of about 4% each year. Data traffic and IP traffic specifically, have grown at an annual rate of 30–50% for over three decades. With tens of thousands of network systems and millions of facilities, automating demand tracking and capacity management is one of the most resource intensive jobs in engineering.
1.2.3.4 Network Redesign
This is the most neglected, and often the most valuable tool available to network engineers. The demand mechanisms cited above are all triggered by events. Capacity augmentation, the most common engineering activity, is triggered when a facility or network element falls below a performance threshold, such as packet discards or blocked calls. Network engineers generally look at those links exceeding the accepted levels and order augmentation or resizing. If a node nears exhaust, either because of port exhaust or throughput limits, engineers order a new node and rearrange the traffic between that node and adjacent ones. In effect they limit the problem and the solution space to a very narrow area, the particular link or node that exceeded a threshold.
Network redesign broadens the scope to an entire region or community. It is performed by systems engineers, not network engineers. It begins with A-Z (A to Z) traffic demand and uses existing topology, link, and element traffic loads as an informational starting point, not as a constraint. In voice networks Call Detail Records (CDRs) are a starting point since they have the calling party (A) and the called party (Z). In IP networks netflow data, coupled with routing information yield the necessary A-Z matrices. Redesigns are performed far too infrequently and the results often reveal dramatic changes in traffic patterns that no one recognized. Express routes, bypassing overloaded network elements, elimination of elements, and rehoming often result in dramatic savings and performance improvements.
1.3 Network Taxonomy
To better understand network operations and engineering some grounding in networks and systems is needed. Networks are best described as communications pathways that have both horizontal and vertical dimensions. The horizontal dimension encompasses the different types of networks which, when operated in collaboration deliver end to end services. The vertical dimension is two tiered. Network elements, which carry user information and critical signaling information, are loosely organized around the OSI seven-layer model, one of the most successful design models in the last 50 years. As a word of warning, I use the terms network system and network element interchangeably. Network system was in wide use when I joined Bell Telephone Laboratories in the 1970s. Network element evolved in the 1990s and is institutionalized in the 1996 Telecommunications Act.
Above the network tier is a set of management systems that look after the health, performance, and engineering of the network tier.
The distinction between network and management systems is almost universally a clear line, as shown in Figure 1.4. Tests used to distinguish between the two systems types are based on how transactions are carried.
Figure 1.4 Network and management systems.
1.3.1 Voice Systems
In the first half of the twentieth century transactions meant one thing, a wireline phone call. A wireline call has six distinct stages.
These six stages are the same whether a call is originated or terminated by a human or a machine. A wide range of technologies has been used over the years in each stage, but the stages are more or less constant.
For voice services we can then distinguish among systems by applying the following tests:
Does the system perform a critical function in one of the six stages of call processing?
If the system is out of service, can existing subscribers continue to place and receive calls?
Network systems when tested yield a yes to the first test and a no to the second. The time frame for applying the tests is important; a reasonable boundary for applying these tests is an hour. A local switching system is the one that gives dial tone and rings wireline phones. If it fails, the effects are immediate. At a minimum no new originations or completions occur because dial tone and ringing are not provided. In severe cases, calls in progress are cut off.
A provisioning system is a counter example. That system is responsible for adding new customers, removing customers, and making changes to existing customers' services. It does not perform a critical function in any of the six stages of call processing. If the provisioning system fails, we simply can't modify a customer's service attributes until the provisioning system returns to service. Existing calls, new originations, and terminations are not affected, so the provisioning system is a management system, not a network system. A second example is billing systems. If a billing system fails on a voice switching system, calls are completed without charge. Unfortunately no one sounds a siren or sends a tweet to let users know the billing system has failed and you can make free calls. The design choice to let calls go free during billing system failure is a calculated economic decision. It is cheaper to let them go free than it is to design billing systems to network system standards. Occasionally losing some revenue is cheaper than building redundant fault tolerant recording everywhere.
But what about the power systems in buildings where communications systems are located? In general network systems operate directly off of DC batteries which are in turn charged by a combination of AC systems and rectifiers. These hybrid power systems are engineered to survive 4–8 hours when commercial AC power is lost. Most central offices have back up diesel generators as well, enabling continuous operation indefinitely, assuming the fuel supply is replenished. Cooling systems fall into the same category. These are systems that do not affect the six stages of voice network systems if they remain failed for an hour. So here is a class of systems that if failed, don't affect calls within our hour time frame, but can affect them after a few hours or possibly days, depending on the season. These systems are in a third category, common systems. This is an eclectic group, covering power, cooling, humidity, door alarms, and other systems that if failed, can imperil the network systems within hours under the wrong circumstances.
1.3.2 Data Systems
The original tiered distinction and design for network and management systems came from the wireline voice network, but it applies to data and mobile networks as well. Consider two common data services upon which we can form our definitions, Internet browsing and mobile texting, or Short Messaging Service (SMS). Browsing is generally performed by a subscriber accessing content on the Internet by sending requests to a set of servers at a web site. The subscriber unconsciously judges the service by the response time, measured from the time the return key is stroked until the screen paints with the response. In the case of SMS, the subscriber has no clear way of knowing when the message is delivered, or if it is delivered at all. However, if a dialog between two SMS subscribers is underway, a slow response or dropped message is likely to be noticed.
For mobile subscribers, many of the network systems that carry Internet service and SMS are common. Our criterion for distinguishing between network and management systems is set by the most demanding data service. Before the introduction of 4G mobile services under the banner of LTE, Long Term Evolution, Internet access was the most demanding service. But LTE, unlike prior mobile technologies, uses Voice over Internet Protocol (VoIP) for voice service. With LTE data (VoIP data) delay tolerances become more unforgiving.
For data systems we can use our voice tests to distinguish among systems by applying the same tests, with minor modifications:
Does the system perform a critical function in the timely delivery of subscriber data?
If the system is out of service, can existing subscribers continue to send and receive data?
The modifier timely was added to the first test. While it was not included in the comparable test for voice service, it was implied. Recalling the six steps of call processing, untimely delivery of any of the functions is tantamount to failure. If you pick up a wireline receiver and have to wait over 10 seconds for dial tone, it's likely one of two things will occur. If you're listening for a dial tone you may grow impatient and just hang up and try again. If you don't listen and just begin dialing, believing the network is ready, you'll either be routed to a recording telling you the call has failed, or you'll get a wrong number. Consider the case of not listening for dial tone before dialing your friend whose number is 679–1148. You could be in for a surprise. Suppose you fail to listen for dial tone and begin dialing. If dial tone is delivered after the 7, the first three digits the switching system records are 911. Now you will have an unplanned conversation with the Public Service Answering Point (PSAP) dispatcher. When Trimline®1
