The Next Generation Internet Protocol IPV6 An Over
The Next Generation Internet Protocol IPV6 An Over
follows: Section 2 describes the current internet host number. It can have up to 125 networks with 16
protocol and the main problems with it. Section 3 million hosts per network. All of these addresses are
covers an overview of the next generation protocol, taken by the world’s big organizations.
IPV6, and discusses its main features and major Class B addresses contains a 14-bit network number,
goals. Finally, section 4 gives a conclusion and a and a 16-bit host number. The number of the
summary. networks which can have this class of addressing
is 16,384, and each network can contain 65,536
CURRENT INTERNET PROTOCOL hosts.
Class C addresses are suitable for the small networks.
The current Internet Protocol is called IPv4, it uses This class has 21-bit of network number and 8-bit
32-bit address fields and has the header format shown host address. This can contain up to 2 million
in Figure 1. networks with 256 hosts per network.
Figure 1: IPV4 Header Format Main Problems with the Current Internet
Version IHL Type of Total Length Protocol
4 bits 4 Services 16 bits The main problems of the IPv4 can be
bits 8 bits divided into two main classes of problems: Address
Identification Flags Fragmentation problems that are related to the limited address space
16 bits 3 Offset and address allocation, and Options problems that are
bits 13 bits related to options required by new applications
TTL Protocol Header Checksum development like QoS. In what follows, I will
8 bits 8 bits 16 bits
summarize the different problems in each class.
Source Address
32 bits
Destination Address Address problems:
32 bits Limited address space: IPv4 has a 32 bit
Options Padding address field which gives up to four
billion addresses. Although this seems
IPv4 protocol organizes addressing by simply like a very large number of addresses,
combining network number and host number. So, in many experts believe that we are nearly
the 32 bits address field the network and the host out of these available addresses. The
under that network is defined. There are three address space of IPv4 has been used up
different kinds of addressing, which are Class A, to a considerable extent. At the end of
Class B, and Class C. As each class of the address January 2008 about 16% was left in the
will support different number of network and hosts IANA pool, i.e. approximately 700
under networks, addresses are assigned to a million IPv4 addresses. There are
network, according to the size of (hosts in the widely quoted and regularly updated
network) the network. estimates which forecast the exhaustion
of the unallocated IANA pool
somewhere between 2010 and 2011
These classes differ in the number of nets and hosts
[10]. One main reason of this is because
as shown in Figure 2:
a significant number of organizations
joining the Internet in the early days
Figure 2: Three main classes of IPv4's addresses
were granted multiple large blocks of
class A addresses. This means that
organization like Apple, HP, MIT,
Stanford etc each have more address
space than China.
Inefficient allocation: Routing is only
practical by separating large blocks of
address space into networks. So the
practical limit is a few hundred million
addresses. In other words, the addresses
are not efficiently assigned and used.
For example, if an organization has a bit
Class A address is the largest host supporting class. more hosts than Class C it will apply for
This class has a 7-bit network number, and a 24-bit Class B address and as a result of this
the most of the host addresses will not Video conferancing. In the current
be used by the organization. To solve version the QoS is low. [6]
this problem a new protocol named Autoconfiguration: IPv4 do not
Common Interdomain Routing Protocol support auto configuration. Making
(CIDR). In this protocol instead of manually configuration of each
assigning a class B address to an host is time consuming and
organization it assign a block of Class B difficult. IPv6 requires that hosts
addresses. But still this is not enough to and routers must support automatic
solve the problem of addressing. configuration, it is not an option.
Routing table size: Routing table size is Typically manager program the
growing so fast has always been a routers with their addresses and
problem. Using two hierarchies to networks then simply plug a host
define the number of connected hosts to onto a network. The host will talk
internet resulted in a huge size routing to the network to get its IP address
table. So it takes more time to process and its routes. It is even possible to
the routing tables. In general the have a network with no routers or
network speed has been increasing servers and still communicate.
faster than router hardware, so extra Using of LANs: By using IPv4 on
performance goes into shovelling LANs there should be a
packets faster rather than handling correspondence between IPv4 and
bigger routing tables. It's not only an MAC. IPv4 provides this by using
issue of routing table size, but also the ARP that utilizes broadcast MAC
dynamic nature of routing updates. layer transmissions [8]. On the
other hand, in IPv6 this is done by
Options related problems: using "neighbor discovery"
Provider selection: By this way method. This method is more
somebody can choose special efficient than ARP when used on
providers for routing the packet. In LANs.
the commercial usage of internet
this take an important role. For
example while sending data one NEXT GENERATION INTERNET
needs a trusted provider. This PROTOCOL: IPV6
feature is not supported in IPv4. [3]
Scalable multicast. In Ip protocols Overview of the IPV6
multicasting is possible only in The next generation Internet Protocol, called IPv4 or
subnets. In some situations like real IPng, uses 128-bit address fields and has the header
time applications it can be possible format shown in Figure 2.
to address different hosts in
different subnets. [5] Figure 2: IPv6 Header Format
Mobility. "plug-and-play": As the
mobility is the target of the IP Version Priority Flow Label
development. The current version 4 bits 4 bits 24 bits
do not support mobile connection. Payload Length Next Hop
Actually you have to configure 16 bits Header Limit
manually. Instead of this there 8 bits 8 bits
should be autoconfiguration to
Source Address
make mobility more efficient. This
128 bits
would mean that your host would
have to get an address from a Destination Address
provider. [6] 128 bits
Real-time flow: If we look at the
development of the Internet, the It is a new version of the Internet Protocol, designed
target is real time data transfer and to take an evolutionary step from IPv4. It was not a
Multimedia. So, the protocol design goal to take a radical step away from IPv4.
should support this applications in Functions which work in IPv4 were kept in IPV6.
an efficient way. For example Functions which didn't work were removed. The
most visible change in IPv6 is that addresses balloon DNS servers could all benefit from IPv6’s
from 32-bits to 128-bits. With such a huge address anycast technology.
space, ISPs will have sufficient IP addresses to Required security protocols. Secure
allocate enough addresses to every customer so that networks will be easier to build and deploy
every IP device has a truly unique address. in an IPv6 because the new IPSec security
Additional address space will also help the core of protocols, ESP (encapsulating security
the Internet by reducing the size and complexity of protocol) and AH (authentication header)
the global routing tables. Although IPv6 doesn’t are built-in and required in IPV6 and not
solve the problems of routing in the Internet, it can add-ons like in IPv4.
help in several areas, reducing the initial size of the
tables and offering a hierarchical address space. Major changes from IPV4 to IPV6
The changes from IPv4 to IPV6 can be primarily
Major goals of IPV6 categorized as the following:
The major goals of IPV6 can be summarized as
follows: Expanded Routing and Addressing
Capabilities. IPV6 increases the IP address
Reduce the total time which people have to size from 32 bits to 128 bits, to support more
spend configuring and managing systems. levels of addressing hierarchy and a much
An IPv6 system can participate in "stateless" greater number of addressable nodes, and
autoconfiguration, where it creates a simpler auto-configuration of addresses.
guaranteed-unique IP address by combining IPV6 addresses are four times the number of
its LAN MAC address with a prefix bits as IPv4 addresses (128 vs. 32). This is 4
provided by the network router [5] Billion times 4 Billion times 4 Billion
Speed up the network, both from a (2^^96) times the size of the IPv4 address
performance and from a deployment point of space (2^^32). This works out to be:
view. IPv6 embodies the lessons learned at 340,282,366,920,938,463,463,374,607,431,7
trying to build high-speed routers for IPv4 68,211,456. It may seem like overkill to
by changing the header of the IP packet to have this many addresses available,
be more regular and to streamline the work However, many visionary individuals
of high-speed routers moving packets across believe that eventually every wired and
the Internet backbone. IPv6 has fixed header wireless computer, cell phone, PDA,
sizes and less number of fields. A side effect household appliance, security camera,
of the redesign of the IP packet header is devices that haven't yet been invented, will
that future extensions to IPv6 are simplified: each have their own unique Internet address
adding a new option to IP can be done [8]. These address are written as:
without a major re-engineering of IP routers o eight colon-separated 16bit hex
everywhere. values, eg
Provide multicast capabilities for high- fe80:0000:0000:0000:02
bandwidth multimedia and fault tolerance 00:e8ff:fe4e:0000
applications. Multimedia applications can o leading zeroes can be omitted from
take advantage of multicast: the each 16bit value,
transmission of a single datagram to fe80:0:0:0:200:e8ff:fe
multiple receivers. Although IPv4 has some 4e:0
multicast capabilities, these are optional and o contiguous zero values can be
not every router and host supports them. replaced by '::',
With IPv6, multicast is a requirement. IPv6 fe80::200:e8ff:fe4e:0
also defines a new kind of service, called A new type of address called a "anycast
"anycast." Like multicast, anycast has address" is defined, to identify sets of nodes
groups of nodes which send and receive where a packet sent to an anycast address is
packets. But when a packet is sent to an delivered to one of the nodes. The use of
anycast group in IPv6, it is only delivered to anycast addresses in the IPng source route
one of the members of the group. This new allows nodes to control the path which their
capability is especially appropriate in a traffic flows.
fault-tolerant environment: web servers and Header Format Simplification. Some IPv4
header fields have been dropped or made
REFERENCES
N.J., IV; Benchoff, P.E. IEEE Network European Parliament, the council, the
Volume: 12 1, Jan-Feb 1998, Pages: 28-33 European economic and social committee
9. Blanchet, Marc & Parent, Florent, IPv6 and the committee of the regions. Brussels,
Primer, 05.04.1999 : 27/05/2008
http://www.viagenie.qc.ca/en/ipv6/tutorial- http://ec.europa.eu/information_society/poli
ipv6-ripe33/html/index.htm cy/ipv6/docs/european_day/communication_
10. The Commission of the European final_27052008_en.pdf
Communities communication to the