TCP/IP Model – Layers and Architecture

The OSI model was a conceptual model, however a more prevalent model which is widely used currently is the TCP/IP model. It was conceived and developed in the 1960s by the US Department of Defense (DoD). It is an abbreviation for Transmission Control Protocol/Internet Protocol. TCP/IP is a compact and smaller version of the OSI model. It has four layers, three less than the OSI model’s seven.

Layers of the TCP/IP model:

1. Network Access Layer:

The lowest and most basic layer of the TCP/IP architecture is the network layer. A network layer is the OSI reference model’s mix of the Physical and Data Link layers. It specifies how data should be physically sent via a network.

This layer is primarily in charge of data transfer between two network devices. This layer’s duties include encapsulating IP datagrams in network frames and translating IP addresses to physical addresses. This layer employs ethernet, token ring, FDDI, X.25, and frame relay protocols.

2. Internet Layer:

The internet layer is the TCP/IP model’s second layer. The network layer is another name for the internet layer.

The internet layer’s primary job is to send packets from any network and ensure that they reach their destination regardless of the route they take.

The protocols used in this layer are:

a. Internet Protocol (IP):

• IP Addressing: This protocol implements IP addresses, which are logical host addresses. The internet and higher levels utilise IP addresses to identify devices and facilitate internetwork routing.
• Communication between hosts: It specifies the path over which the data will be sent.
• Data Encapsulation and Formatting: The data from the transport layer protocol is accepted by an IP protocol. An IP protocol guarantees that data is securely delivered and received by encapsulating it in a message known as an IP datagram.
• Maximum Transmission Unit: The limit imposed on the size of the IP datagram by the data link layer protocol is known as fragmentation and reassembly (MTU). If the size of an IP datagram exceeds the MTU unit, the IP protocol divides the datagram into smaller units to allow it to transit over the local network. Fragmentation can be performed by either the sender or an intermediary router. All of the pieces are reconstructed at the receiver side to produce the original message.
• Direct delivery occurs when an IP datagram is transmitted via the same local network, such as a LAN, MAN, or WAN. When the source and destination networks are on different networks, the IP datagram is transmitted indirectly. This is done by routing the IP datagram through various devices such as routers.

b. Address Resolution Protocol (ARP):

ARP is a network layer mechanism that determines the physical address based on the IP address.

The following two words are mostly connected with the ARP Protocol:

• ARP request: When a sender needs to know the physical address of a device, the ARP request is broadcast to the network.
• ARP reply: Every device connected to the network will receive and execute the ARP request, but only the recipient will recognise the IP address and respond with its physical address in the form of an ARP reply. The physical address is added to the recipient’s cache memory as well as the datagram header.

c. Internet Control Message Protocol (ICMP):

It is a method used by hosts or routers to transmit datagram issue notifications back to the sender.

A datagram is routed from router to router until it arrives at its destination. If a router is unable to route data due to unexpected circumstances such as disabled connections or network congestion, the ICMP protocol is used to notify the sender that the datagram is undeliverable.

An ICMP protocol primarily employs two terms:

• ICMP Test: The ICMP Test is used to determine whether or not the destination is accessible.
• ICMP Reply: ICMP Reply is used to determine whether or not the destination device is responding.

The ICMP protocol’s primary role is to notify issues, not to solve them. The sender bears responsibility for the adjustment.

Because the IP datagram includes the addresses of the source and destination but not of the router to which it is sent, ICMP can only transmit messages to the source and not to the intermediate routers.

3. Transport Layer:

The transport layer is in charge of data dependability, flow control, and correction as it travels through the network. It employs two protocols: User Datagram Protocol and Transmission Control Protocol.

a. User Datagram Protocol (UDP):

It offers connectionless service and end-to-end transmission delivery. It is an unreliable protocol since it detects faults but does not describe the error.

The problem is discovered by the User Datagram Protocol, and the ICMP protocol notifies the sender that the user datagram has been corrupted.

UDP is made up of the following fields:

• The source port address is the address of the application software that generated the message.
• The destination port address is the address of the application software that will receive the message.
• The total length is: It specifies the total amount of bytes in the user datagram.
• Checksum: A checksum is a 16-bit field that is used to detect errors.

The UDP protocol does not specify which packet is lost. UDP simply includes a checksum; it does not carry any data segment ID.

b. Transmission Control Protocol (TCP):

It offers complete transport layer services to apps. It establishes a virtual circuit between the sender and receiver and remains active during the communication.

TCP is a reliable protocol since it identifies errors and retransmits broken frames. As a result, it guarantees that all segments are received and acknowledged before the transmission is deemed complete and a virtual circuit is destroyed.

TCP splits the whole message into smaller units known as segments at the sending end, and each segment carries a sequence number that is necessary for reordering the frames to create the original message.

TCP gathers all of the segments and reorders them depending on sequence numbers at the receiving end.

4. Application Layer:

In the TCP/IP architecture, the application layer is at the top. It is in charge of high-level protocols and representation concerns. The user can interact with a computer program through this layer.

When one application layer protocol wishes to interact with another, it sends its data to the transport layer. Except for those that interface with the communication system, no application can be installed within the application layer.

Protocols used in the application layer are:

a. Hypertext Transfer Protocol (HTTP):

HTTP is an abbreviation for Hypertext Transfer Protocol. This protocol enables us to access data through the internet. It sends data in the form of plain text, audio, and video. It is characterised as a hypertext transfer protocol because it is effective in a hypertext context with quick leaps from one document to another.

b. Simple Network Management Protocol (SNMP):

SNMP stands for Simple Network Management Protocol. It is a framework for controlling internet-connected devices utilising the TCP/IP protocol stack.

c. Simple Mail Transfer Protocol (SMTP):

SMTP stands for Simple Mail Transfer Protocol. A Simple mail transmission protocol is the TCP/IP protocol that enables e-mail. This protocol is used to transfer data to a different e-mail address.

d. Domain Name System (DNS):

DNS is an abbreviation for Domain Name System. An IP address is used to uniquely identify a host’s access to the internet. However, individuals prefer to use their names rather than their addresses. As a result, the system that maps a name to an address is known as the Domain Name System.

e. TELNET:

It is short for Terminal Network. It connects the local computer to the distant computer in such a way that the local terminal seems to be a terminal on the remote system.

f. File Transfer Protocol (FTP):

FTP is an abbreviation for File Transfer Protocol. FTP is a common internet protocol that is used to transfer data from one computer to another.

g. Secure Shell (SSH):

Secure Shell is an abbreviation for Secure Shell. It is a terminal emulation program comparable to Telnet. Because of its capacity to maintain the encrypted connection, SSH is chosen. It establishes a secure session through a TCP/IP connection.

h. Network Time Protocol (NTP):

NTP is an abbreviation for Network Time Protocol. It is used to synchronise our computer’s clocks to a single standard time source. It is extremely beneficial in circumstances such as bank transactions.

Advantages of TCP/IP Model:

• It assists you in establishing a connection between various sorts of computers.
• This is not reliant on the operating system.
• It supports a wide range of routing protocols.
• It allows organizations to communicate with one another over the internet.
• The TCP/IP architecture features a client-server design that is extremely scalable.
• It can function on its own.
• A variety of routing protocols is supported.
• It is capable of establishing a link between two computers.

Disadvantages of TCP/IP Model:

• TCP/IP is a difficult model to set up and maintain.
• TCP/IP has a greater shallow/overhead than IPX (Internetwork Packet Exchange).
• In this approach, the transport layer does not ensure packet delivery.
• It is difficult to replace a protocol in TCP/IP.
• It does not have a clear distinction between its services, interfaces, and protocols.

Difference between OSI Model and TCP/IP Model:

Summary:

In this article, we cover all of the different layers of the TCP/IP model, and we also take a look at all of the different protocols present in each of these layers. We also take a look at the advantages and disadvantages of the TCP/IP Model, in addition to the differences between the TCP/IP and the OSI Models.