IPv6: Internet Protocol Next Generation

Internet and network development lately has been making the Internet Protocol (IP)-based networking is the backbone of TCP / IP is quickly becoming obsolete. If two decades ago, the IP is connecting millions of computers, then the current IP should be able to connect billions of computers. So far, Internet and network TCP / IP the other designed to support the use of such simple-distributed file transfer (ftp), electronic mail (e-mail), remote access (remote access) by using TELNET. But today, due to the explosion of popularity of the World Wide Web in recent years, the Internet has grown into multimedia. At the same time in the company's network has shifted from e-mail and use simple file transfers to the environment client / server complex. These developments have made exceeding the capacity of IP-based network to supply the necessary services and functions. An environment like the Internet requires support for data traffic in real-time and security functions. This need is very difficult to be fulfilled by the IP version 4, or often called IPv4. Development of IPv6, IP next generation, or IPng (IP next generation) - recommended in the IETF meeting in Toronto on July 25, 1994 - backed by a shortage of IP addresses that currently has a length of 32 bits, due to tissue growth explosion. IPv6 is a new version of IP which is the development of IPv4. IP address shortage IP address is a row along the 32-bit binary numbers, which are used to identify hosts on the network. IP addresses are given uniquely to each computer / host which is connected to the internet. Packet that carries data, loaded the IP address of the sender's computer data, and IP address of the destination computer, then the data is sent to the network. Packet is then sent from router to router based on the IP address, go to the computer. All host / computer that is connected to the Internet, differentiated only by IP addresses, so it is clear that there can be duplication. So IP addresses are distributed by several organizations that have authority over the distribution of IP addresses, such as the Inter-NIC (Network Information Center). At first, with 32-bit address is deemed sufficient to be shared on the host. But with the tremendous growth of the Internet in recent years, there are fears this will end IP address, due to exceptional demand. During this inter-NIC provides the IP address by using the concept of class. That is, throughout the 32-bit IP address is divided into 2, namely the network and the host, the Inter-NIC only manage the network only. In other words, when given an IP address in an organization, the Inter-NIC only give part of its network only, while the rest, giving the IP address on each host submitted on the organization. In IPv4 there are 3 types of classes, depending on the size of the host, namely class A (all 24-bit host part, the IP address can be given to host 16.7 million), class B (all 16-bit host part host = 65 534) and class C (host part all 8 bits = 254 hosts). Network administrator to apply for the class type scale based on a network that dikelolanya.Konsep this class have the advantage of: the management of information routes do not require all 32 bits, but quite simply the network only, so large that the route information stored in the router, becomes small. Concept Classes Once the network address is obtained, then the organization can freely give the address of the host on each host. But surely not all addresses obtained can be used. In fact many unused. For example, organizations that are connected to the internet with a number of hosts as many as 1,000 pieces, because the number of that host more than 256, then selects the class B. Consequently only a small proportion than 65536 addresses acquired is used. Meanwhile, the remaining addresses can not be used by other organizations. As a result, the current shortage of IP addresses, particularly the class B. To overcome this, taken way, instead of giving a class, then given some smaller classes. For example in 1000 hosted the above case, instead of class B, class C is given by 4 units. Order to get the IP in a more efficient distribution. But instead, this has led to the emergence of another problem, namely swelling of routing information. Routing information Swelling Routers on the internet, serves to choose the appropriate route to send the packet, by referring to the IP address contained in the packet. For routers that must manage the network address of the host and the relationship with the interface in a routing table (routing table). In this routing table, each network address, given a single item. If the router is on the LAN, the routing information that must be managed is not too much. But the main Internet routers routing information that must be managed to be very much. Since each added a network on the Internet, the amount of data in tables is greatly increased as well. As mentioned above, if an organization has multiple address class C, then the same number of routing data to be managed. If organisani the class B, then only need 1 routing data only, consider that the organization has 4 class C, then the routing of data that must be managed as many as 4 pieces. As a result of this data and then a very long memory of the router. To reduce the amount of data routing, then introduced the concept of CIDR (Classless Routing InterDomain). This technology eliminates the concept of class, aiming for the provision of address space in an organization can be more flexible. For example, the organization that received 4 fruit class C, given the lower 10 bits of 32-bit address as the host address. So the network address that must be managed to be one that is part of a 22 bit remainder. By using this method, routing information can be greatly reduced. However, one can use the CIDR is only a continuous address space only. In addition, CIDR can only be applied to the new address given, whereas if you want to apply to the old address so the address must be given back (given a new address.) With CIDR, the swelling problem of routing information can be overcome, but because the Internet continues to grow rapidly, CIDR does not solve the problem fundamentally.
(1) With a unified, addresses in the routing table only to 1 only, but need additional new field about the length of bit parts network. Automation of various settings IPv4 Address on essentially static on the host. Usually given sequentially on the host. For example when using a computer note, if connected to a network, the IP address used is the IP address on the network. If connected to another network, the IP address must be set back to the IP address on the network. On a dialup connection, since the IP addresses keep changing it must be set back. It is currently above this can be done automatically by using DHCP (Dynamic Host Configuration Protocol). However, it is actually not included in the specification of IPv4, an optional additional functions besifat. In fact, DHCP is only limited to the new OS, so that DHCP must be loaded individually. Therefore, it remains necessary to address a flexible delivery mechanism at the IP layer. Equal one characteristic of the Internet is the use of shared networks, especially backbone. Although the bandwidth (bandwidth) provider of 1.5 Mbps, and so did the routing of them, but not necessarily that we can always use all the bandwidth. If at that time there are other users who use that line, then the bandwidth can we use was reduced as much as that used by the user. In other words there is no guarantee of bandwidth that can be used at any given time. Despite using a large enough line to the provider, when it will send or receive voice or image, but can not be certain to take place smoothly before accomplished. Or in the extreme, when we do a TV conference in the LAN, all of a sudden there are other users who send huge data, then the picture and sound received / sent will falter. Therefore needed a new system that can meet the new usage. Security In order to use the Internet to expand into areas of business and many other fields, the function of security is one of the key. Consequences of the Internet as a public network, can not be denied the possibility of data seen during the trip. Or have prevented access that could damage, such as hackers or crackers. Function security can be realized in various layers. Although there has been a lot of security at the application layer such as for electronic mail, or WWW, still required the function of security at the network layer. In IPv4 itself does not have these security functions. Changes to IPv6 Then what has changed with this IPv6? The main improvement is on: the expansion of address space (IP address), simplifying the header of the packet, Plug & Play, Security function. Each of these improvements are intended to be responding to the growth of the Internet, increasing reliability, and ease of use. Address Expansion As explained above, such a rapid Internet growth in recent years has resulted in scarcity of IP addresses. The biggest change in IPv6 is the extension of the IP address of 32 bits in IPv4 to 128 bits. This is a 128 bit address space by eliminating the concept of the continuous class. It also made changes in the way of writing IP addresses. If the IPv4 32 bits divided into 8 bits each separated it with "." and write with a decimal number, for example, 150.7.7.250. So on IPv6, 128 bits are separated into 16 bits each, where each section is separated by ":" and written by hexadesimal.Contohnya, "4FE5: 2F21: 3512:77 BB: AF23: 3201:55 AA: 2F33". In addition, not only enlarge the address space, but also introduced multilevel structure, for routing management made easy. In CIDR (Classless InterDomain Routing) routing tables is reduced by combining into one routing information from an organization. In IPv4, not at all concerned with the relations between organizations and countries. Meanwhile, on IPv6, some organizations with the same provider, or have geographic ties, linked and mirrored on the routing. In other words, if several organizations are in a one provider at the time of the IP address attempted to address these could be in the same address space. For example there are two organizations on a provider, the two organizations close the distance when viewed from the network. So in giving the address, aligned to the top of the address is not much different as far as possible. Thus, outside providers, routing information from both these organizations can put together. In this way, it's been done division of IP address space in IPv6. And that occupy much space is the address for providers and addresses for the region. each occupies one eighth of the available address space. Then when it was still about 70% that has not been defined, which is left as a reserve for granting a new address. Next, let's see-story structure in IPv6 addresses this by looking at examples in the address to the provider. First of all along the 128-bit address is divided into several fields which can change the length. If the first 3 bits of address are "010", then this is a space for providers. Meanwhile, the next n bits are registry ID is a field that indicates the place / institution that provides the IP address. For example, the IP addresses provided by the InterNIC, the field becomes "11,000". Furthermore, m the next bit is the provider ID, while the next bit o Subscriber ID to distinguish the organizations listed on these providers. Then p next bit is the Subnet ID, which marks a collection of hosts that are connected in the network topology of the organization. And the q = 125 - (n + m + o + p) is the last bit Interface ID, which is the IP address that marks the host contained in groups which have been marked by the Subnet ID. Subnet ID and Interface ID is freely given by the organization. Organizations are free to use the remaining p + q bits of the IP address in providing an IP address within the organization after receiving a 128 - (p + q) initial bits of an IP address. At that time, administrators of these organizations may divide into the sub-network and host bits in the appropriate length, if necessary can also be made more structured again. Because of the long bit on the provider ID and subscriber ID can be changed, then the address given on the provider and the number of IP addresses that can be given by the provider to the user can freely given according to need. In IPv6 routing control part is called prefix in the address field, which can be considered equivalent to a network address in IPv4.
3 Type the address provided IPv6 addresses can be divided into 3 types. The first, called a unicast address, used to communicate one on one, by appointing a single host. Then multicast is used to communicate an opponent a lot. This is to designate the host of the group. Then the last is a new function in the IPv6 anycast address, which refers to the host of the group, but the packet is sent only on one host only. On a host are not always given only one address of the third type above address, but could have given a few addresses. For example, having once Unicast addresses, link local address, and anycast addresses. Unicast Address In unicast addresses, addresses that are assigned glogal such as address to the provider, the geographical address. In addition, a link local address and the site local address. Each of the Link-local address and site-local address, included in the IPv6 address space that takes each about 1 / 1024 of the available address space. Link Local Address is the address used within one link only. What is meant here is a link local network connected to each other on one level. Address is automatically created by hosts who have not received the global address, consisting of 10 + n-bit prefix that starts with "fe80" and fields along the 118-n bits indicating the number of hosts. Link Local Address is used in the delivery of an IP address automatically.
Local Site Address being equivalent to the private address, limited use in the site only. This address can be given freely, provided unique within that site, but could not send the packet to the destination address is outside of the site. Multicast Address While Multicast Address first set the address to a host group. Then if any packet is sent to the address, the packet will be sent to all hosts in the group. Multicast Address is the IPv4 is defined as class D, while in the IPv6 space is 8 bits in his first start with "FF" is provided for multicast addresses. This space is then subdivided again to determine the range of validity. Then Blockcast address in the host IPv4 address is defined as "1", the IPv6 multicast is already included in this Address. Blockcast address for communication in the same segment are separated by a gateway, as well as the multicast address ranges are sorted based on destination. Anycast address In IPv6 was added a new address type that does not exist in IPv4, the anycast address. In this type of address, an address given on multiple hosts, to define a collection of nodes. If any packet sent to this address, the router will send the packet to the nearest host that has the same anycast address. In other words, the owner of the router packet handed goal of the most "suitable" for the delivery of packets. The use of anycast address this example of some of servers that provide services such as DNS (Domain Name Server). By providing the same anycast address on servers, if any packet sent by the client to this address, the router will choose the nearest server and send the packet to the server. Thus, the load on servers can be distributed merata.Bagi anycast address is not provided specific rooms. If for some hosts are given a same address, the address is regarded as an anycast address.

On IPv6 Packet Structure In line with the expansion of the address to 128 bits, then the IPv6 packet structure had been improved as well. In addition, the field is seldom used in IPV 4 was removed replaced with fields that support real time communications and others. In this packet header pendesignan, strived for header processing cost becomes small. For example, start and end address to be needed in each packet. While the IPv4 header when the packet is fragmented, there are fields to store the sequence between the packet. But the field is not used when the packet is not fragmented. IPv6 header consists of two types, the first, namely the fields required by each packet called a basic header, while the second is a field that is not always necessary in a packet called extension headers, and headers are defined separately from the basic header. There is always the basic header on every packet, while the additional headers inserted only as needed between the basic header to the data. Additional headers, is currently defined in addition to use when a packet is broken, also defined the functions of security and others. This additional header, placed after the basic header, if needed some header then this header will be connected a chain starting from the basic header and ends on the data. Routers only need to process the smallest header that needed it, so that processing time becomes faster. The result of this improvement, although the basic header size enlarged from 20 bytes to 40 bytes but the number of field is reduced from 12 to 8 pieces only.
 Structure of the IPv6 basic header Setting address automatically When I first connect the host to the Internet, users need to be setting up an IP address, netmask, routing and others. However, the current number of Internet users continues to grow and even be able to connect from home, so it is not practical when it is necessary for setting up knowledge about the network as the gateway IP address and others. In IPv4 even in the presence of DHCP, the parameters needed for setting up has to be automated. In connection with using a PPP or LAN built with Windows-95/NT already commonly used. But DHCP is a function of option / option in IPv4, so not all systems, and OS support it. In certain systems still several parameters must be set manually. In IPv6 functions for setting up automatically provided by default. In other words in this whole machine IPv6 stay connected on the cable network and immediately be connected to the Internet. on these automatic settings, provided 2 ways depending on the usage of address, namely the automatic settings and statefull stateless. Both ways are selected based on the tendency of administrators. Stateless auto setting is a way that makes the process of managing at least. No need to provide a server for managing and sharing an IP address. Simply setting up the router only. Hosts that are connected in a network of routers that exist on the network, obtained from the address prefix of the network. Then add the host bits pattern obtained from the unique information to the host, then make the IP addresses along the 128-bit and make it as the IP address of the host. In the information unique to this host, used among other MAC addresses of network interfaces. In this stateless auto setting behind the ease of management, on an Ethernet or FDDI because of the need to provide at least 48 bits (by MAC address) on one network, has the disadvantage of poor efficiency in the use of addresses. The second way is automatic settings statefull's how the management of strictly in terms of range of IP addresses provided on the host by providing a server for managing state of the IP address. This method is similar to the way DHCP in IPv4, despite having to maintain servers and address, but since IP addresses can be used effectively, this method has the characteristics of administrators to manage in detail as to structuring the address in the organization and others. When do the settings automatically, the required information between routers, servers and hosts are ICMP (Internet Control Message Protocol) which has been expanded. In the ICMP in IPv6, including IGMP (Internet Group Management Protocol) is used to multicast in IPv4. Security Functions Currently in the transaction by using the internet has become a reality, security at the time of the communication is not negotiable. Currently, the method by using S-HTTP (Secure HTTP) for sending credit card numbers, or personal data to encrypt, or encrypt e-mail with PGP (Pretty Good Privacy) has been used in general. However, the above is securiti offered by the application. In other words if you want to use these functions we must use the application. If you need security on communications without depending on specific applications will require security functions at the TCP or IP layer. And IPv6 support encrypted communications and authentication at the IP layer. By having a security function on the IP itself, then it can do things like packet sent from a particular host entirely encrypted. In IPv6 for authentication and encrypted communication using an extended header called AH (Authentication Header) and the encrypted payload, called ESP (Encapsulating Security Payload). In the second communication requires encryption or one of the header is added. Security function that is used at the application layer, such as the S-HTTP use SSL as a method encripsi, while on PGP uses IDEA as a method encripsinya. While the use of key management in a certain way as well. Instead, the IPv6 is not defined in a certain way in encripsi and key management method. So be flexible to use any method. This is known as the SA (Security Association). Function Security in addition to the use of IPv6 on encrypted communication between a pair of hosts, can also perform encrypted communications between networks by encrypting the packet by the gateway from 2 networks that make such communication. Label Flow and Real Time Process The header of the packet in the IPv6 flow label field (flow-label). This label, used to ask for pakect certain treatment is given by the router while in transit. In other words, these labels can signal the type of packet. For instance on sound or moving images (motion picture) as far as possible be transferred as soon as possible even though the quality is slightly reduced. While e-mail or the WWW is more need to accurately than on the nature of real time.
The table on the IPv6 Flow Label Label Categories 0 Uncharacterized Traffic 1 "Filler" traffic (e.g., netnews) 2 Unattended data transfer (e.g., e-mail) 3 Reserved 4 Attended bulk transfer (e.g., FTP, HTTP, NFS) 5 Reserved 6 Interactive traffic (e.g., Telnet, X) 7 Internet control traffic (eg, routing protocols, SNMP) 8-15 Realtime communications traffic, non-congestion-controlled traffic Routers manage priorities and resources such as communication capacity or ability to process, based on this flow label. If the IPv4 packet is treated the same throughout, then at this IPv6 with different treatment to each packet, depending on the contents of the packet, can be realized applicative communications. Cover Internet Protocol version 4 which is the foundation of the Internet, has almost approached the final frontier of his ability, and IPv6 which is a new protocol has been designed to replace IPv4. The primary motivation for change was the lack of IPv4 addresses that long, only 32 bits only. In addition, IP is a protocol that is very old who are unable to support the need for secure communications, flexible routing and traffic control data. These reasons that encourage the emergence of IPv6, the next generation Internet protocol. IPv6 address that has a huge capacity, supporting the preparation of the address in a structured, which allows the Internet continues to grow and provide new routing capabilities not found in IPv4. IPv6 anycast address type that can be used to efficiently route selection. In addition, IPv6 is also complemented by the use of addresses in a local mechanism that allows the realization of a Plug & Play installation. IPv6 also provides a platform for new ways of Internet usage, such as support for data streams in real-time, provider selection, host mobility, end-to-end security, or automatic configuration.