IPv6.. Advantages and Structure

1. IPv6.. Advantages and Structure

2. Current IPv4 Internet • Evolved out of a 1960s US DoD experiment. • The core workings of the Internet lie in the ‘routing’ of data packets tagged with IP addresses (similar to telephony numbering). • Currently a 32-bit addressing scheme is utilized in IPv4. • The commercial Internet started formation between 1986-1991. • Over 25 years old with no or little changes in the protocol/communication method. • Because of rapid growth demands there was no or little foresight into the potential and exponential growth of the Internet. • This could eventually lead to critical issues in the Internet world.

3. Current Internet Trends • Broadband. – Content rich, video, voice, digital photography and music. • Utilization of the Internet for ‘mission-critical’ communications – IPSec VPNs, interoffice voice traffic, etc • Interconnectivity of non-PC based equipment. – IP phones, building automation systems. • End-to-end communications/peer-to-peer (P2P). – No more server-client model. – Sensors, network file sharing and games. • Ubiquitous communications. – Hotspots, mobility, 3G, PDAs.

4. IPv4 Obstacles • Network Address Translation (NAT) complications. – Effective temporary solution but introduces complexities in network design and transparency for end-to-end communications. - Breaks the globally unique address model. - Breaks address stability. - Restricts ‘always-on’ and ‘plug-and play’ environments. - Breaks peer-to-peer model. - Breaks some security protocols. - Breaks some QoS functionalities. • No host based and end-to-end security. • Limited IP addresses allocations. – IPv4 cannot scale considering current population and economic growth. • Poor mobility solution e.g. triangular routing issue. • Poor Quality of Service (QoS) solution and routing efficiency for the current Internet infrastructure. • Basically showing its age.

5. IPv4 NAT Restrictions • Analogous to PABX/private phone extension vs. publicly reachable phone number. • Reliant on features of NAT on router, firewall, gateway. • Cannot scale well and service and functionality limiting.

6. IPv4 Routing Issues • Large global routing table size. • Explosion of the Internet and poor routing administration has lead to the exponential growth of the global routing table. Currently at over 130k prefixes. • Routing performance. • At every hop, the router has to check and verify the header checksum. This will increase processing time and degrade router performance. • Fragmentation of IP packets. • Fragmentation is also done by routers. IP packets might need to be fragmented several times. • End-node has no way of knowing the number of fragments. • If any fragments are lost or delayed, the datagram will be discarded. • This will also effect routing performance.

7. Why IPv6? An Introduction • Limitations of IPv4 and future extensions are addressed. • Worked on for the last 7 years by the IETF (Internet Engineering Task Force). • Addressing schema is based on a 128-bit address length. – 6 digit to 8 digit telephone numbers analogy. – Virtually unlimited global IP address: – Population of the world: 6,000,000,000 (sometime ago!) – IPv4 addresses: 4,294,967,296 – IPv6 addresses: 340,282,366,920,938,463,463,374,607,431,768,211,456

8. Note IPv6 allows 7 x 1023 IP addresses per each square meter on earth surface (land and water)

9. Why IPv6? An Introduction ../contd. • Plug and play function ‘automagically’ supported. - Serverless auto-configuration and reconfiguration via neighbor discovery and router advertisements. Improved operational efficiency. - Ability to interconnect very small devices in large numbers, with mobility and security functions, etc. • Inherent IPSec security features. - Provides built-in end-to-end security. • QoS is a standard and being further developed. • Designed to support mobility –mobile IP (MIP6) - The ability to ‘roam’ from network to network without loosing end-to-end connectivity. • Home address : Maintains connectivity. • Care-of address : Maintains route-ability. • Other extensions but not limited to: - Multicast/anycast, routing efficiency, and other future applications. - Future proof options.

10. Why Larger Addresses Range? • Perhaps the most important aspect of IPv6 is the larger IP address allocation available. • The Internet is still expanding. - 320 million users in 2000, and more than 500 million user in 2005. • Increase in user based connected devices. - 405 million mobile phones in 2000, over 1 billion in 2005 e.g. 3G. - 1 billion cars in 2010 and 15% expected to use GPS, locality based applications, and sensors. • Potential growth for new non-PC Internet based applications. - TV, camera, radio, fridge. • Growing geopolitical and economic drivers. - Driving for a new global knowledge based economy. - Sustainable economic growth by remaining competitive and dynamic. - Stanford University, MIT, Xerox, and Apple each have more IPv4 addresses than the whole of China. IPv6 addresses readily available for all.

11. IPv6 Readiness: Transition Process • Transition between today’s IPv4 Internet and the future IPv6 one will be a long process in which both protocols will coexist. • IPv4 replacement will be soft and gradual – no Y2K ‘flag-day’. - Estimated 2030-2040 for IPv4 to be fully replaced/extinct. Current seamless transition techniques: Gateway translators, IPv6/IPv4 tunneling, 6to4, IPv6 tunnel broker services. • Current preferred transition method is IPv4/IPv6 dual-stack. • Past transition examples. 16-bit application to 32-bit application transition period in1995.

12. IPv6 Strengths Revisited • There are no major changes in the fundamental workings of IP between IPv4 and IPv6. • IPv6 has however addressed current IPv4 ‘shortcomings’, included additional features and is designed to be future-proof.

13. XDA PDA File Sharing IP Phone 3G Phone Sensors Web Camera Car Navigation Home Appliances Some IP Applications and Usage

14. New Internet IPv6 Business Models

15. End-to-end Secure Communications

16. IP Packet Structure

17. IPv4 Header Format

18. IPv4 Header Fields

19. IPv4 Header Fields ../contd.

20. IPv6 Header Design • Recognizable and streamlined. • Reduce common-case/flow processing cost of packet handling. • Keep bandwidth overhead low in-spite of increased size of header (due to 128-bit address length). • Flexible and extensible support: - IP Protocol Type in IPv4 replaced as optional Extension Headers. - Extension Headers can be chained together. • Fixed header length. No need for Padding Field. • Fewer fields in header allows for faster processing: - No header checksum calculation. - Upper layer checksums mandatory. • 64-bit alignment header/options. • Efficient option processing - Only processed when present and mostly processed at destination. • No fragmentation in the network. - Reduce router load and layer 3 switching of IP

21. IPv6 Header Format

22. IPv6 Header Fields

23. IPv6 Extension Header *Extension Headers can be ‘chained’ and include hop-by-hop, routing, fragmentation, authentication, and encrypted security payload options.

24. IPv6 Extension Header ../contd. • Basic header simplified for ease of processing. • Additional information carried in extension headers e.g. hop-by-hop options, routing header, fragment header, etc. • Next Header field says what type of header follows – e.g. fragment header, TCP, ICMP, etc.

25. IPv6 Extension Header ../contd.

26. IPv4 Header Format Changes

27. IP Header Changes IPv4 Header 20 octets, 12 fields, including 3 flag bits and fixed maximum number of options. IPv6 Header 40 octets, 8 fields, and unlimited chained Extension Headers.

28. Functionality Comparison

29. IP Header Format Issues & Improvement • Checksum in IPv4 header format will calculate only header checksum. – Computation done if there are changes in the header. As the TTL value decrements at every hop, re-computation of this field is done. – Decreases efficiency of the router. • Options and Padding fields are checked at every router hop. – As with checksum, this increases router processing time per-packet. – Taking into consideration the exponential growth in bandwidth, this will lead to degraded router performance. • In IPv6 there are no header checksums. Transport and data-link layers already perform ‘checksumming’. The removal of this feature improves faster IP packet processing. • No Options field in IPv6 and replaced with Extension Header. The removal of the Options Field results in a fixed length 40-bit byte IP header. • No fragmentation procedure by routers. – IPv6 fragmentation and reassembly is an end-to-end function. – With path MTU discovery in IPv6, only the source host performs the fragmentation process (unlike in IPv4 any host in the routing path can perform fragmentation). – The removal of this procedure will speed up IP forwarding in routers.

30. Overlay Tunnels •Tunnels can be made over multiple routers and networks. •Tunnels can be terminated by routers or hosts.

31. Configuring Tunnels (Cisco) 1. Create Interface Tunnel 2. Configure IPv6 Address for Point to Point 3. Configure tunnel source as local router IPv4 address 4. Configure tunnel destination as remote router IPv4 address 5. Configure tunnel mode ipv6ip

32. 1500 byte packet Packet too big/MTU=1480 1480 byte packet Packet too big/MTU=1280 Successful packet delivery 1280 byte packet IPv6 MTU and Path MTU Discovery • In IPv4, every router can fragment packets if needed. Can cause inefficiency in throughput and packet reordering. • In IPv6, routers do not perform packet fragmentation -done by end-nodes. Sending hosts carries out path MTU discovery steps to determine the largest packet size supported on the route. Packet Too Big message used and will never go below the IPv6 minimum MTU of 1280 bytes.

33. ICMPv6 • Similar concepts and functionalities as in IPv4. • IGMP protocol which manages multicast group memberships has been incorporated into ICMPv6. • ARP/RARP functions are now ICMPv6 functions. • There are two classes of ICMP messages: error and informational. • Error messages: – Destination unreachable – Packet Too Big – Time Exceeded – Parameter Problem • Informational messages: – Echo request/Reply – Router Solicitation/Advertisement – Multicast Listener Discovery • RFC2463 outlines the processing rules of ICMPv6.