Internet Protocol and IP Forwarding
Solving the problem of scaling with the Ethernet, and allowing the support of the other types of LANs, as well as the linking of one point to the other, are the basic reasons for which the Internet Protocol, the IP was developed. In many cases, we consider that supporting the universal connectivity is the central idea for designing the internet protocol. In this, anything can connect to everything else. The system does not want to result in unmanaged large forwarding tables, that is why it allows scaling the enormous size. In the earlier times, the size of the nodes which appeared was 10^9. But it was required that it should have the size of more than this.
The internetworks of the basic networks were the Internet Protocols in earlier times. For the unaware, the basic networks were the LANs. But, the things have changed now. The users ignore this and take the internet as one large network.
There is a global mechanism which the IP provides for supporting the universal connectivity. It is done for addressing and routing. This also makes the delivery of one host to the other possible.
IPv4 and IPv6
IPv4 is the IP addresses which use 4 bytes, i.e. (4bytes*8= 32 bits). They are also the part of the IP header. The header, in this case, follows the Ethernet header. Staying with the packet of one hop is the work which the Ethernet header does. For the entire journey of the packet, the IP header always stays with it.
The networks, be it IPv4 or IPv6, these both can be divided into the several other networks and their part as well as the host. The parts of the network are called the prefixes, and the hosts are named as the remainder. Dividing the network according to the first few bits, the legacy information and mechanisms are designed for this purpose.
TCP / IP Model and Their Comparison is also important for the examinations point of view.
Unlike the Ethernet addresses, the IP addresses are assigned administratively. In the earlier times, the prefix of the Class B network was assigned from the IANA. It is Internet Assigned Numbers Authority. After this, turning the assignment of the host portion in the form of appropriate for the local site. The services which the administrative assignment are extensive. These are not only the endpoint identification as well as the locator. They include embedding the information of the location.
The definition of the Class A/B/C was first coined in 1981 in RFC 791. RFC 988 was used to introduce Class D, and it was added in 1986. The initial bits of Class D is 1110. These addresses are used multicast. It involves sending a set of recipients to every packet of IP.
The host bits and the division of the network are dynamic. That means these can be modified, in the form of positions. The address and the levels of the network are changed. The IP header does not carry the network and the host division point. Negotiating the division point of negotiating the next_hop forwarding information is the work done by the routers.
Network Number is the network portion of an IP address. The other name of the network portion is network prefix and network address. The network prefix is used to make all the decisions of forwarding the packets. For the denotation of the network prefix, the host bits are set to zero. Along with this, the result of the address is followed by the number of network bits in the address.
The hosts which are lying on the same network address are called to be on the same network of IP. These are to be located on the same LAN and at the same time. When two hosts are sharing the same network address, then it is said that they can reach each other on a direct basis. The underlying LAN is the medium through which they can communicate. If they fail to contact to each other, then we experience connection failure. The rule has many consequences. One and the most prominent is that only the network nodes are to be looking for the site of the packet. All the hosts share the same network prefix on the same physical LAN. This is why they are a part of the same network of IP. This is the most prominent reason why the LAN is set for its division into several IP networks.
Sizes of the packets
The maximum packet size is with each LAN technology. For instance, Ethernet holds the packet size of 1500 bytes. Whereas, if we talk about the once-competing Token Ring, it had a maximum size of 4KB.
Ethernet is the concept on which the today’s world has been standardized on. Most of the times, Ethernet packet-size limits. But the IP does not have the same case. The best effort system is IP. There is an absence of the retransmissions and acknowledgements in the IP-layer. The packets are shipped off, and it reaches there. The IP-layer and the other networks involve representing connectionless networking. It does not need to maintain the information that goes from end to end point. The techniques which it follows is the delivery of the packets like in the form of giant LAN. The next layers, the TCP is given the responsibility of the creation and the maintenance of the connections. The alternate ways other than the connectionless networking of doing the things are the connection-oriented internetworking. Each of the types, connectionless and connection-oriented networking. The former is more reliable as compared to the other.