cover

CONTENTS

Foreword

Preface

Acknowledgments

Authors, Contributors, and Reviewers

1 IPv6 Drivers in Broadband Networks

1.1 IPv6-Based Services

1.2 Broadband Access Models

1.3 Summary

References

2 IPv6 Overview

2.1 IPv6 Protocol Basics

2.2 Summar

References

3 Deploying IPv6 in Cable Networks

3.1 Cable Network Elements

3.2 Cable Networks Today

3.3 Summary

References

4 IPv6 Deployment in DSL, ETTH, and Wireless Networks

4.1 New Remote Access Architecture for IPv6

4.2 DSL Networks

4.3 Ethernet Networks

4.4 IEEE 802.11A/B/G Wireless Networks

4.5 Summary

References

5 Configuring and Troubleshooting IPv6 on Gateway Routers and Hosts

5.1 IPv6 Support on Gateway Routers

5.2 IPv6 Support on Windows XP, Windows Vista, and Windows Server 2003 and 2008

5.3 IPv6 Support on Linux

5.4 IPv6 Support on MAC OS X

5.5 PPPv6 Support on MAC OS X

5.6 IPv6 Support on Solaris

5.7 Troubleshooting IPv6 on GWR and Hosts

5.8 Summary

References

6 Configuring and Troubleshooting IPv6 on Edge Routers

6.1 IPv6 Configuration on the Edge Router

6.2 Summary

References

7 Configuring and Troubleshooting IPv6 on Provisioning Servers

7.1 IPv6 Support on DHCP Servers

7.2 IPv6 Support on DNS Servers

7.3 IPv6 Support on TFTP Servers

7.4 IPv6 Support on AAA and RADIUS Servers

7.5 Troubleshooting IPv6 on an ER and on RADIUS AAA Servers

7.6 Summary

References

8 Conclusion

8.1 IPv6 Addressing Considerations

8.2 IPv4-IPv6 and IPv6-IPv6 Interworking

8.3 Subscriber Logging

8.4 Recovery Options

8.5 Summary

Appendix A IPv6 Case Study

Appendix B DHCPv6 Message Types and Option Codes

Index

title

I dedicate this book to my parents for everything they have done for me; I could never repay them. To my wife: for her love, support, and patience throughout the writing of this book. To my lovely children: Asad, Aashir, and Zeerak, you are my inspiration and my pride and joy. I love you all very much.

I would also like to dedicate the book to all the people suffering in this world and to those people who work endlessly to help others and to make the world a better place.

Adeel Ahmed

I dedicate this book to all who are suffering and going through tribulations, and to those who are working to relieve those in pain and suffering. I also dedicate this work to my loving parents, brothers, and two lights of my life: Bahira and Fatimah!

Salman Asadullah

FOREWORD

The Internet is becoming a utility with an estimated 1.5 billion users, commonly referred to as “netizens,” around the world. This large user base is surpassed only by the 3.5 billion mobile (cell) phone users on the planet. Approximately 10% of cell phones in use today are “smart phones,” which also provide Internet services. The Internet backbone is quite robust; however, the last mile of the access layer is made up of a fragmented delivery system ranging from very low speed to high-speed (1 Gbps) connections. To put things in perspective, a 1-Gbps connection will allow downloading of a James Bond movie in about 20 seconds. High-speed download allows better use of Internet resources than does live streaming, which is resource intensive due to time constraints on packet delivery and reordering.

Broadband provides the Internet with an opportunity to become a robust utility similar to the TV networks. It is interesting to note that TV networks were designed with enough capacity to match viewer expectations for quality of experience. The next stage in the evolution of broadband access is to move from asymmetric to symmetric provisioning, thus allowing consumers to download and upload at the same speed. IPv6 provides a large address capacity and will be ideal for a commodity addressing scheme that will enable two-way always-on, Internet services. This scenario will signify the most important shift in adoption of the Internet, by empowering users to become “full-time residents” instead of simply sporadic in-and-out consumers. The evolving Internet offers a bright future by transforming casual users in TV broadcasters, reporters, and remote entertainers, and by creating new jobs and providing unprecedented opportunities that traditionally required expensive support infrastructure. The deployment of IPv6 has become an issue of strategic importance for many national economies. Telecom operators and Internet service providers (ISPs) are key players in facilitating the deployment of IPv6 on broadband access networks amid the obvious complexities of coexisting with or replacing widely deployed IPv4 services. Evidently, telecom operators and ISPs have to take steps to ensure a workable transition strategy that involves transparent interoperability and integration of mature and advanced applications over both IPv4 and IPv6, this strategy will enable a combination of services that will allow service providers to explore and exploit richer services offered by IPv6 during a potentially long transition from IPv4. This will also lead to new business models that will generate return on investment without waiting for the ubiquitous deployment of IPv6.

The IPv4 address space is virtually depleted, with just over 14% capacity remaining, and is expected to run out by the end of 2010. It is therefore anticipated that the use of IPv6 will gain momentum and the end user will drive differentiated services, achieving returns not only in investments but also in service innovations and flexible communication solutions. Solutions for integrating and deploying both IPv4 and IPv6 services are mature and available to service providers.

The authors of this book have the necessary technical expertise and experience to identify the challenges and to propose recommendations and solutions of great value to a world made of heterogeneous and widely un-interoperable networks designed using private addressing schemes that inhibit end-to-end applications and services. Their extensive involvement in standardization bodies such as the IETF and knowledge spread at the customer level in the design and deployment of IPv6 networks are of paramount value to readers, who can gain first-class knowledge to empower them to tackle the transition to IPv6 with greater confidence.

Welcome to the new two-way IPv6-based broadband access to the Internet!

LATIF LADID
President, IPv6 Forum

PREFACE

This book is meant to be used as a guide by network engineers and architects while deploying IPv6 in their broadband networks. Service providers world¬wide are looking for ways to expand their networks and to meet the scalability requirements of the growing number of always-on devices connected to the Internet. IPv6 is currently the only solution available to meet these challenges and to enable service providers to scale their networks and provide new services to customers.

The focus of the book is on many of the challenges faced by service providers today and how IPv6 addresses these issues. Chapter 1 covers drivers for IPv6 in broadband networks to give readers an idea of why they should be looking at deploying IPv6 in their networks. Chapter 2 provides an overview of IPv6 protocol basics and provisioning.

Chapters 3 and 4 are concentrated on IPv6 deployment techniques in various broadband networks, such as cable, DSL, ETTH, and wireless. Comparisons are drawn between IPv4 and IPv6 deployment models, and similarities and differences between the two are discussed in detail.

The concluding four chapters deal in detail with the configuration of different network components in an IPv6 broadband solution, and provide guidelines on debugging and troubleshooting IPv6-related issues. This will help readers configure and troubleshoot problems when they deploy IPv6 in their networks.

Appendix A has a real-life SP case study of Free’s IPv6 broadband deployment. Free is the second-largest French ISP with more than four million broadband subscribers (ADSL and FTTH).

This book is intended for those network engineers and architects who are contemplating deploying IPv6 in their broadband networks. The book contains detailed information as to how IPv6 can be deployed in service provider broadband networks. Different IPv6 deployment models, configurations, and troubleshooting guidelines are discussed to help readers understand the challenges faced by SP in deploying IPv6 in broadband access networks.

Adeel Ahmed Salman Asadullah

ACKNOWLEDGMENTS

First and foremost, the authors acknowledge Benoit Lourdelet for his major contributions to Chapter 4 and for reviewing the remaining chapters and providing valuable feedback. The authors also thank Latif Ladid and Patrick Grossetete most sincerely for honoring us by endorsing the book and by writing the foreword and the endorsement notes.

The authors appreciate the hard work of all the reviewers, whose valuable insights and feedback improved the book significantly. Our special thanks go to Abe Martey for making available his vast knowledge of networking technology and for his rigorous review of the text. Abe’s prior experience as the author of two networking technology books, IS-IS Network Design Solutions and Troubleshooting IP Routing Protocols, both published by Cisco Press, and his detailed comments and feedback, were great assets and made our work more valuable for the audience.

We would also like to thank all the people whom we have worked with on the IPv6 front at Cisco, vendors, operators, IETF, other standards bodies and platforms, and industry at large, to make IPv6 a reality.

We especially thank George Telecki, Michael Christian, Angioline Loredo, and all the staff at Wiley for helping us drive this book to completion.

Special Acknowledgments by Salman Asadullah: I would like to thank a few people who although not directly related to this work have helped me through the journey of life in one way or an other: Muhammad ibn ’Abdullah and his companions, Mohammad Asadullah, Shakila Siddiqui, Salah Uddin Ayubi, Salik bin Saddina, Hamaza Yousf, Sohaib Webb, Usama Canon, Anwar Awlaki, Abdul Sattar Edhi, Dr. Ali Metwally, Dr. Magda Mohsen, Imran Asadullah, Bahira Metwally, Kamal Siddiqui, Khalid Raza, Nasir Kamal, Romana Khan, Mike Quinn, Syed Khurram, John Selden, Iqbal Ahmed Khan, Abdul Mateen Hashmi, Adeel Ahmed, Fawad Asadullah, Zulfiqar Ahmed, Rasheed Uddin, Robert Santiago, and Himanshu Desai.

Special Acknowledgments by Adeel Ahmed: I would like to acknowledge a few people who have especially influenced my life and have provided guidance and inspiration: Muhammad ibn ’Abdullah and his companions, Sanjeeda Ahmed, Aziz Ahmed, Shakila Khanum, Khairunisa Begum, Shaikh Suhaib Webb, Shaikh Saad Hassanin, Natasha Ahmed, Najia Ahmed, Farzana Khan, Umar Saeed, Shaukat Khalil, Nasir Ali, and Salman Asadullah.

ABOUT THE AUTHORS

This book is written by Cisco Certified Internetwork Experts (CCIEs) who have been working with various customers worldwide on IPv6 deployment, standard bodies, and technical forums since 2002. The authors have worked with several development teams within Cisco in driving IPv6 implementation on Cisco products and influencing the IPv6 technology direction through their work with standards bodies. They have written and contributed to several white papers, design guides, and IETF RFCs and drafts on deploying IPv6, and have provided numerous trainings and seminars on this topic. The authors’ combined 22 years of Internetworking industry experience and 12 years of experience in working with IPv6 brings valuable knowledge and expertise to the book.

Adeel Ahmed, CCIE No. 4574, is a Technical Leader in Cisco’s Advanced Services group. He has been with Cisco Systems for over 10 years. His areas of expertise include access/dial, broadband cable, and IPv6. He has worked with major cable MSOs in North America, EMEA, and ASIAPAC in designing and troubleshooting cable networks. He has written several white papers and design guides used by customers, sales teams, and Cisco engineers in deploying multiservices over cable networks. He has also coauthored and contributed to IETF RFCs and drafts.

Ahmed has represented Cisco at industry technical forums such as IETF, CableLabs, NCTA, SCTE, Networkers, APRICOT, NAv6TF, and Global IPv6 Summit. He holds a bachelor’s and a master’s degree in electrical engineering.

Salman Asadullah, CCIE No. 2240, is a Technical Leader at Cisco Systems and holds honorary positions at APRICOT, NSP, and IPv6 Forum. Recognized as an expert within Cisco and by Cisco’s customers and industry as a whole, he has been designing and troubleshooting large-scale IP and multiservice networks for over 13 years. He has represented Cisco in industry panel discussions and technical platforms such as Networkers, APRICOT, NANOG, SANOG, IETF, and IPv6 Forum events.

Asadullah influences technology and product directions and decisions within Cisco business units and in the Internet community. He has produced several technical documents, white papers, and articles and has coauthored and contributed to IETF RFCs and drafts. He is a coauthor of two networking technology books: Cisco CCIE Fundamentals: Network Design & Case Study, and PDIO of the IP Telephony Networks, both published by Cisco Press. PDIO of the IP Telephony Networks is a best-seller, with over 13,000 copies sold to date. He holds a B.S. in electrical engineering from Arizona and a M.S. in electrical engineering from Kansas.

ABOUT THE CONTRIBUTORS

Benoit Lourdelet is an IPv6 Senior Product Manager at Cisco Systems. He helps to drive the range of Cisco products that support IPv6. Benoit has been a key player in the deployment and architecture of the first IPv6 broadband networks across the globe. As a known IPv6 expert, he continues to influence the architecture designs of next-generation IPv6 broadband networks. He has over 15 years of experience in designing and operating telecommunications networks. He has also worked with both manufacturers and service providers in developing Internet exchange points such as PARIX and international backbones. He has contributed toward many IPv6 technical papers and IETF RFCs and drafts. He is regular speaker on IPv6 topics both at Cisco events and at IPv6 international conferences such as Cisco Networkers, AFNOG, and IPv6 Forum events. He holds an M.S. in chemistry and an M.S. in computer science from institutions in France.

Alexandre Cassen works as a software architect at Freebox, the R&D lab of Free.fr. He focuses on large-scale protocol and software design and implementation of value-added services. Most of his time is dedicated to developing software for IPTV and VoIP networking components. He enjoys learning new networking protocols and technologies.

ABOUT THE REVIEWERS

Abe Martey, CCIE No. 2373, works in the focused technical support group in Cisco Systems, where he provides expert support on Cisco’s high-speed router platforms and other SP core networking technologies to major worldwide SPs. Abe is the author of IS-IS Network Design Solutions and coauthor of Troubleshooting IP Routing Protocols, both published by Cisco Press. Abe has been with Cisco for over 10 years, during which he has held several positions ranging from technical support and technical and product marketing. Prior to Cisco, Abe was a network engineer at Sprint, where he worked in Sprint’s Managed Router Network Group and also as a support engineer in the early days of the Sprintlink IP Network. Abe has an MS in electrical engineering and is an active member of the IEEE.

Srinivasa Neppalli, CCIE No. 6370, works as a Network Consulting Engineer in the Advanced Services Broadband Team at Cisco Systems. He has been with Cisco since 1999. He holds CCIE certifications in R&S and service provider tracks. He has been providing consulting service for service providers and cable MSOs for the last seven years. He focuses on IP backbone technologies and his expertise includes routing protocols, MPLS applications, multicast, and IPv6. He holds a master’s degree in electrical engineering.

Roy Boos is a Network Consulting Engineer in the Cisco Advanced Services Group. He has been with Cisco for over 10 years. His areas of expertise include network management, SNMP, broadband cable, and provisioning. His customers have included major MSOs in North America and ASIAPAC. He has provided detailed audits of their provisioning systems and authored several best practices white papers on monitoring recommendations for the 7246VXR, uBRlOK, and 7600 via SNMP. Roy holds a bachelor’s degree in computer engineering.

Michael Reekie is a Network Consulting Engineer in the Cisco Systems Advanced Services (AS) team. Michael has been involved in the testing and deployment of the Cisco Network Registrar (CNR) DHCP and DNS servers, and the Broadband Access Center family of products since joining Cisco in 1999 as part of the American Internet acquisition. In 2001, he played a key role in performance testing of the redesigned CNR DHCP protocol engine. Later he created an in-depth training program for engineers tasked with supporting CNR in the Cisco Technical Assistance Center (TAC). Michael joined Cisco Advanced Services in 2006 and has been supporting Tier 1 Cable MSO customer’ provisioning data and voice services since then. He received his bachelor’s degree in 1993 from Boston University.

1

IPv6 Drivers in Broadband Networks

With the exponential growth of the Internet and an increasing number of end users, service providers (SPs) are looking for new ways to evolve their current network architecture to meet the needs of Internet-ready appliances, new applications, and new services. Internet Protocol Version 6 (IPv6) is designed to enable SPs to meet these challenges and provide new services to their customers.

The life of IPv4 was extended by using techniques such as network address translation (NAT) and other innovative address allocation schemes. However, the need for intermediate nodes to manipulate data payload while employing these schemes posed a challenge to peer-to-peer communications, end-to-end security, and quality of service (QoS) deployments. IPv6 also addresses fundamental limitations in the IPv4 protocol that renders the latter incapable of meeting long-term requirements of commercial applications. Besides its inherent capabilities to overcome the aforementioned limitations, IPv6 also supports an address space quadruple of that of IPv4, by supporting 128-bit instead of 32-bit addresses (RFC3513). The huge IPv6 address space will enable IPv6 to accommodate the impending worldwide explosion in Internet use. IPv6 addressing provides ample addresses for connecting consumer home/Internet appliances, IP phones for voice and video, mobile phones, web servers, and so on, to the Internet without using IP address conversion, pooling, and temporary allocation techniques. IPv6 is designed to enhance end-to-end security, mobile communications, and QoS, and also to ease system manage¬ment burdens, as the protocol is still evolving, with some of its capabilities still a work in progress by the Internet Engineering Task Force (IETF).

As the number of broadband users increase exponentially worldwide, cable, digital subscriber line (DSL), Ethernet to the home (ETTH), wireless, and other always-on access technologies, along with IPv6’s huge address range, long-lived connections to servers, and permanent prefixes for home appliances, give more flexibility to SPs. Specifically, the addressing capacity of IPv6 has made it valuable to SPs, most of which are rolling out IPv6 support in their networks or aggressively evaluating its potential and value for service delivery. Outside the United States, IPv6 adoption is being promoted on a national level, and countries such as Japan, Korea, China, India, and some European countries have taken lead roles in moving from testing and evaluation to actual deployment in broadband services and applications.

Cable, DSL, ETTH, and wireless services are the main broadband technol¬ogies that are widely deployed. In this book we discuss key aspects of IPv6-enabled broadband networks and explore differences from IPv4 deployments.

1.1 IPv6-BASED SERVICES

Until recently, IPv6-based services were considered primarily as differentiators that allowed SPs to exploit the large address space available in IPv6 for future growth planning and as a competitive advantage. However, as IPv6 has become more popular and familiar, SPs are adopting the protocol not only to offer new services to their customers but also for provisioning and managing a large number of network devices and applications. Governmental interest and promotion by means of incentives and favorable legislation is also a major driver for the growing adoption of IPv6 in SP and enterprise networks.

SPs in densely populated regions such as Asia and Europe are at the forefront of adoption and integration of IPv6 into their networks to address the increasing numbers of broadband subscribers and the greater scaling of their networks. For example, Nippon Telephone and Telegraph (NTT) in Japan is currently offering dual-stack commercial services to asymmetric digital subscriber line (ADSL) and fiber to the home (FTTH) subscribers. Dual-stack devices are capable of forwarding both IPv4 and IPv6 packets; hence, these FTTH and ADSL users can access the services with either an IPv4 or an IPv6 address, or both. In this deployment model, subscribers are offered /64 dedicated IPv6 prefixes but generally receive only a single static or dynamic IPv4 address.

Additionally, some SPs are offering integrated IPv6-based multicast and voice over IP (VoIP) service in addition to existing IPv4-based services, and are taking advantage of the larger IPv6 address space and other useful features. The multicast services consist of several video and audio streams available simultaneously to broadband subscribers. The content providers store the content based on users’ interests and send this content as multicast streams to broadband subscribers. Today, with IPv4 service offerings, generally a single device attached directly to a gateway router (GWR) at the customer’s premises receives the multicast stream. In similar IPv6 offerings, multiple devices may be attached to the GWR, each receiving a different stream at the same time.

For example, in Japan cable and satellite TV are not very popular, and users expect to receive video content through traditional broadcast TV programs. This provides an opportunity for content service providers to generate addi¬tional revenue by offering content not available through TV to broadband subscribers at reasonable prices. The content provider may multicast several channels of video and audio, and broadband subscribers will join various multicast groups of interest to receive content. Disney movies are an example of a video stream, and an audio stream could be karaoke. In this regard, this service offering is similar to a cable TV subscription.

In North American and the Asia Pacific region, IPv6 is being adopted primarily by large cable multiple systems operators (MSOs) to address growing subscriber-base and IP-enabled devices. Currently, cable MSOs use RFC1918 address space for cable modems and set-top-box management. RFC1918 provides 16 million addresses under the 10.0.0.0/8 prefix, plus 1 million addresses under 172.16.0.0/12 and 65,000 addresses under 192.168.0.0/16 prefixes. Factually, address utilization efficiency decreases with hierarchical topologies (see the HD ratio in RFC1715 and RFC3194), so the 9.8 million cable modems and set-top boxes could easily exhaust all 16 million RFC1918 private addresses. The exhaustion of IPv4 private address space among the cable MSO community has become a catalyst in the definition and standardi¬zation of the data over the cable service interface specification (DOCSIS) 3.0 standard, which introduces support of IPv6 in cable networks. Now, cable MSOs are planning to deploy IPv6 to manage this large number of cable modems and set-top boxes.

SPs are also offering IPv6 services over wireless links using 802.11-compliant WiFi hot spots. This enables users to take notebook PCs and PDAs along with them and connect to the Internet from various locations. One of the potential benefits of this service flexibility may be downloading digital pictures from a mobile phone with a digital camera to a home storage server.

Figure 1.1 depicts an end-to-end SP network with some of the most commonly deployed access broadband technologies, a core network and back-end provisioning, and network management servers. The access layer of the network features different terminologies for provider edge (PE) devices in the various access models; however, in all cases, PE devices function similarly by acting as head-end devices. The head-end devices connect to multiple downstream customer premises devices using different encapsulation techni¬ques, protocols, and methodologies. In this book we explore the integration and deployment of IPv6 in the broadband access segment as well as in provisioning and network management servers. We also highlight some commonly used techniques for enabling the network core for carrying IPv6 traffic. Once the SP has enabled IPv6 in an access broadband, core, and on back-end provisioning and management servers, it may consider enabling IPv6 on a per application basis. For example, the SP may choose to manage the network devices using IPv6 as a first application.

1.2 BROADBAND ACCESS MODELS

With the exception of cable MSOs, two access models are prevalent in most access broadband technologies, such as ETTH, DSL, and wireless local area networks (WLANs). In the first model the network access provider (NAP) and the network service provider (NSP) are owned and operated by the same entity. This is referred to as the ISP-operated model. In the second model the NAP and the NSP are owned and operated by two separate entities. This is known as the wholesale deployment model.

Figure 1.1 Typical end-to-end SP access broadband network.

c01f001

Generally, NAP is the entity that provides last-mile access services, and NSP is the entity that provides Layer3 services to customers. NAP and NSP operations are explained in detail in Chapter 4. Cable MSO architectures are very different from ETTH, DSL, and WLAN architectures and are covered in detail in Chapter 3.

1.2.1 ISP-Operated Deployment Model

In an ISP-operated model, a PE router [also known as an edge router (ER)], which is owned by the broadband SP, assigns the IPv4 address to the GWR. This deployment model is depicted in Figure 1.2. Assignment of the IPv4 address is done either by DHCPv4 or by static configuration. The IPv4 traffic is sent from the GWR to the PE using point-to-point protocol over Ethernet (PPPoE) (RFC2516), point-to-point protocol over ATM (PPPoA) (RFC2364), or routed bridged encapsulation (RBE) access methods, which are valid encapsulation candidates to offer IPv6 connectivity in a variety of access SP designs. Methods that are PPP based can leverage the IPv6 extensions to the authentication authorization and accounting (AAA) framework and the remote authentication dial-in user service (RADIUS) protocol (RFC3162) and fit in well with current IPv4 deployment models.

When deploying IPv6 in this framework, the same encapsulation techniques are used and the same behavior is achieved by ER assigning the IPv6 address to the GWR. The assignment of an IPv6 address is done either by using stateless address autoconfiguration (SLAAC), DHCPv6 using AAA and RADIUS, or static configuration. Depending on the agreement between the broadband SP and the user, this IPv6 address assignment could have a prefix length of /64 or shorter. The ER and GWR are upgraded to dual-stack routers in order to support both IPv4 and IPv6. The only scenario when GWR does not need to be in dual-stack mode is when the host behind the GWR is running a PPPoE client and the IPv6 address is assigned directly to the host by the ER router, with the GWR acting only as a bridge.

Figure 1.2 ISP-operated deployment model.

c01f002

1.2.2 Wholesale Deployment Model

In the wholesale deployment model the NAP only provides network access to users; it does not deal with addressing issues. NAP transports the user’s IPv4 traffic from the GWR to the NSP ER [which acts as an L2TP network server (LNS)] by various means, such as Layer2 circuits [virtual local area networks (VLANs)], ATM permanent virtual circuits (PVCs), or frame relay virtual circuits (VCs), or by using other common encapsulation techniques (these PPP sessions can further be bundled be into an L2TP tunnel). The NAP’s L2TP access concentrator (LAC) initiates a L2TP tunnel to the NSP’s ER router. The LNS assigns IPv4 addresses to the devices (routers, PCs, etc.) located behind the GWR using DHCPv4 with AAA and RADIUS or by static configuration. Once the L2TP tunnel comes up, all traffic is forwarded to the NSP’s ER over the L2TP tunnel. In a nutshell, the NSP’s ER terminates these circuits and acts as a Layer3 gateway for all these users. For this reason, the NSP is actually responsible for assigning IP addresses to users. Figure 1.3 illustrates the wholesale deployment model operation.

In this model, if the NSP wants to provide content via multicast, it has to replicate that content to all subscribers. To preserve network resource one tries to do the replication as close to users as possible. In the case of an NSP reaching its customers through a wholesale NAP, the closest Layer3 device to the users is the NSP ER that terminates the virtual circuits. This means that the NSP will have to replicate the packets for all the virtual circuits with the users who requested it, and flood the NAP infrastructure with multicast replications. This is not a problem for the NSP; however, the wholesale NAP now has its network flooded with duplicate packets. This is not optimal, and it limits dramatically the capability of the wholesale NAP to scale support for the NSP’s IP multicast service. Chapter 4 covers in detail new deployment models supported by IPv6 for addressing such scaling challenges.

Figure 1.3 Wholesale deployment model.

c01f003

When integrating IPv6 support in this model, the GWR and NAP’s LAC are upgraded to dual-stack routers to support both IPv4 and IPv6. It is not necessary to upgrade the GWR to dual-stack capability when the host behind the GWR is running a PPPoE client, and the IPv6 address is assigned directly to the host from the NSP ER, with GWR acting only as a bridge. The NAP’s LAC initiates a L2TP tunnel to the NSP’s ER to forward traffic received from the GWR. The NSP ER assigns IPv6 addresses to the devices (routers, PCs, etc.) located behind the GWR using SLAAC, DHCPv6 using AAA and RADIUS, or static configuration. The GWR may receive a shorter than /64 prefix, depending on the NSP’s policy and the customer’s requirements. In this case the NSP’s ER router is also upgraded to dual-stack status to support both IPv4 and IPv6.

1.3 SUMMARY

IPv6 enables SPs to offer new services as well as to enhance current services with a focus on servicing endpoints. IPv6 services may range from network addressing support for home appliances to peer-to-peer communication, such as Internet gaming, music and video file sharing, and IP telephony. With sufficient address space in the foreseeable future for several billion subscribers, appliances, and applications, IPv6 is the gateway to the future of the next-generation Internet. It is interesting to note that the world’s population is projected to be 10 billion in 2050 and that up to 34 billion IPv6 addresses will be available to be assigned to each person as well as to many of the planets appliances, automobiles, buildings, cameras, control units, embedded systems, home networks, medical devices, mobile devices, monitors, output devices, phones, robots, sensors, switches, and VPNs. Thus, IPv6 holds the key to the success of the next-generation Internet.

REFERENCES

1. S. Asadullah and A. Ahmed, “IPv6 in Broadband,” Cisco Systems, Inc. Packet Magazine, Fourth Quarter 2004.

2. S. Asadullah, A. Ahmed, C. Popoviciu, P. Savola, and J. Palet, “ISP IPv6 Deployment Scenarios in Broadband Access Networks,” RFC4779, January 2007.

3. B. Lourdelet, “Application Note: IPv6 Access Services,” Cisco Systems, Inc.

4. B. Lourdelet, “Application Note: DHCPvö,” Cisco Systems, Inc.

5. Cisco Systems Inc., tutorial on IPv6 basics: “The ABC of IPv6.”