It was introduced in one of the oldest wide area network - ARPANET. It is a widely used architecture because most of the computers and networks are configured to follow this suite and it is open to all.
The TCP/IP reference model is a 4 layered model. The layers from top to bottom are Application Layer, Transport Layer, Internet Layer and Host-to-Network Layer. A data in a computer is going to the network from the Application Layer to the Host-to-Network Layer through the intermediate layers. Data from the network is getting to an application in the reverse order.
Application Layer consists of certain higher level protocols like TELNET (for remote login/ to get a virtual remote terminal), FTP (File Transfer Protocol), SMTP (Simple Main Transfer Protocol), HTTP (Hyper Text Transfer Protocol), DNS (Domain Name System) protocols etc. The protocols in this layer will produce messages smaller in size to become compatible with the Transport Layer for an outward data flow and join the data messages to the original data for the application in case of inward message.
Transport Layer consist the protocols like TCP (Transmission Control Protocol) and UDP (User Datagram Protocol). The reliability and the error checking of the data are managed here. This layer is responsible for the communication between two computers. ...
The two versions of IP are IPv4 and IPv6.The lowest layer is the Host-to-Network Layer and this layer has the direct contact with the networking hardware. The data in this layer will be in the form of frames or byte series. Few protocols like PPP (Point to Point Protocol) are available in this layer. One of the most useful services of TCP/IP is the naming service it provides. Information sharing and exchange on an Internet work become easy and comfortable through this. The most widely used naming services are NIS (Network Information Service) and DNS (Domain Name System). NIS is used to provide information that has to be known throughout the network, to all machines on the network. There will be NIS server to keep all the information for the NIS. In case of a UNIX system, the configuration files handled by NIS are /etc/group, /etc/hosts, /etc/netmasks, /etc/passwd, /etc/protocols, /etc/rpc, and /etc/services. For setting up NIS, we have to set up both the NIS Server and NIS Clients.
DNS is used for naming a host uniquely on a network, especially on the Internet. In this IP addresses of the computers or other interfaces are mapped to certain names with a typical naming strategy. Each group of similar computers is included in same domain and the domains are created in a hierarchical manner e.g. for the top level domains are .com, .net, .org etc. The details of the naming are stored in the DNS Server as resource records. There are two types of mapping files. The forward mapping file is used to convert names to its IP addresses and reverse mapping record is used to convert IP addresses to its symbolic names. Format of a resource record entry is as follows.
Name (variable length) type(16bits) class(16bits) ttl(32bits) data
For DNS configuration in a UNIX system, we