Table of Contents
The TCP/IP protocol suite, also known as the Transmission Control Protocol/Internet Protocol, is a comprehensive collection of protocols. It operates through multiple layers, with higher-layer protocols relying on the support of lower-layer protocols. Initially, the model comprised four layers:
- Host to Network Layer
- Internet Layer
- Transport Layer
- Application Layer
The diagram below illustrates a comparison between the OSI and TCP/IP models, showcasing the associated protocols.
|
|
Comparison of OSI model and TCP/IP model
|
The TCP/IP model closely resembles the OSI reference model in structure, although it condenses the seven layers of OSI into four layers:
- The Application layer in the TCP/IP model aligns with the Session, Presentation, and Application Layers of the OSI model.
- The Transport layer in the TCP/IP model corresponds to the Transport Layer in the OSI model.
- The Network layer in the TCP/IP model aligns with the Network Layer in the OSI model.
- The Host-to-Network layer in the TCP/IP model corresponds to the Physical and Data Link Layers of the OSI model.
Functions of the Layers of TCP/IP Model
1. Host to Network Layer:
- This layer operates as a fusion of protocols from the physical and data link layers, supporting all standard protocols within these layers.
- It acts as the bridge between the higher-level protocols and the physical network, ensuring seamless communication by integrating the functionalities of both the physical and data link layers.
2. Network Layer:
- Also known as the Internetwork Layer, it houses the Internetworking Protocol (IP) responsible for transmitting data from source to destination.
- The Internetworking Protocol (IP) is a connection-less and unreliable protocol, offering a best-effort delivery service. There is no error checking in IP; instead, it relies on underlying layers for data transmission.
- IP divides data into packets or datagrams of the same size. Each packet is independent, allowing it to be transported across different routes and potentially arriving out of order at the receiver.
- Being connectionless, IP packets find the best possible path to reach the destination without a predefined setup between the sender and receiver.
- Due to the lack of guarantees about data delivery, IP is considered an unreliable protocol. Packets may get dropped along various routes during transmission.
- Despite its unreliability, IP is not weak or useless; it efficiently provides only the necessary functionality for transmitting data, ensuring maximum efficiency.
-
The Network layer is associated with four other protocols:
- ARP
- RARP
- ICMP
- IGMP
- These protocols contribute to the overall functionality of the Network layer, addressing various aspects of data transmission and communication.
i. ARP – Address Resolution Protocol:
- Utilized to determine the physical address of a device on a network when its logical address is already known.
- The physical address, a 48-bit identifier imprinted on the NIC or LAN card, corresponds to the Internet Address, commonly known as the IP address, which uniquely and universally identifies the device.
- In essence, ARP facilitates the translation between the logical and physical addresses, enabling seamless communication within the network.
iii. RARP – Reverse Address Resolution Protocol:
- Employed by a device within a network to discover its Internet address when it possesses knowledge of its physical address.
- In practical terms, RARP serves as a mechanism for a device to obtain its logical (Internet) address based on the known physical address. This is particularly useful in certain network configurations and scenarios.
iii. ICMP – Internet Control Message Protocol:
- Functions as a signaling mechanism to notify the sender of issues with datagrams that arise during transit.
- Primarily utilized by intermediate devices, such as gateways, when they encounter problems like a corrupt datagram.
- In the event of an issue, ICMP enables the intermediate device to send a message back to the sender, providing crucial information about the encountered problem. This mechanism aids in efficient network troubleshooting and maintenance.
iv. IGMP – Internet Group Message Protocol:
- Serves as a mechanism facilitating the transmission of the same message to a designated group of recipients.
- In practical terms, IGMP is essential for enabling efficient communication within multicast groups on a network. It ensures that messages are delivered to the intended audience, optimizing network performance for group-based communications.
3. Transport Layer
Transport layer protocols are accountable for efficiently transmitting data between processes on one machine to their corresponding processes on another machine. Within the transport layer, three primary protocols play distinct roles:
- TCP
- UDP
- SCTP
I. TCP – Transmission Control Protocol
TCP, or Transmission Control Protocol, is a reliable and connection-oriented protocol. This means that a connection is first established between the sender and receiver before data transmission begins, ensuring a secure and ordered exchange.
TCP strategically organizes the data it receives from the upper layer into segments. Each segment is assigned a sequence number, a crucial element used at the receiving end for the accurate reordering of data. This sequencing mechanism enhances the reliability and integrity of data transmission, ensuring that information reaches its destination in the intended order.
II. UDP – User Datagram Protocol
UDP, or User Datagram Protocol, is a straightforward protocol designed for direct process-to-process transmission. It operates as an uncomplicated and connectionless protocol, particularly suitable for applications that do not necessitate flow control or error control.
Unlike TCP, UDP does not establish a connection before data transmission and does not offer the same level of reliability. Instead, it simply appends essential information such as port address, checksum, and length to the data received from the upper layer. This minimalist approach makes UDP a lightweight option for scenarios where simplicity and low overhead are priorities over guaranteed delivery and error correction.
III. SCTP – Stream Control Transmission Protocol
SCTP, or Stream Control Transmission Protocol, is a recently introduced addition to the transport layer of the TCP/IP protocol suite. This protocol stands out by combining the distinctive features of both TCP and UDP.
Primarily employed in applications such as voice over the Internet, SCTP boasts a considerably broader range of applications. Its versatility makes it well-suited for scenarios where a reliable and message-oriented approach is essential, offering advantages in both reliability and flexibility. The integration of features from TCP and UDP positions SCTP as an adaptable solution for diverse communication needs.
4. Application Layer
The Application Layer serves as a consolidation of the Session, Presentation, and Application Layers from the OSI model. It is responsible for defining high-level protocols essential for various applications, including File Transfer (FTP), Electronic Mail (SMTP), Virtual Terminal (TELNET), Domain Name Service (DNS), and more. This layer acts as the interface between software applications and the underlying network, ensuring seamless communication and interoperability across diverse services.
Addressing in TCP/IP
Addressing in TCP/IP refers to the process of assigning unique identifiers to devices within a network to facilitate communication.
The TCP/IP protocol suite involves 4 different types of addressing:
-
- Physical Address
- Logical Address
- Port Address
- Specific Address
1. Physical Address
- The Physical Address, also known as the link address, is the most basic level of addressing.
- It is localized to the network to which the device is connected and is unique within that specific network.
- Found at the data link layer, the physical address is typically embedded in the frame for communication.
- Often referred to as the MAC (Media Access Control) address, it consists of 6 bytes (48 bits) and is permanently imprinted on the Network Interface Card (NIC) of the device.
- The size of the physical address may vary depending on the network type; for instance, Ethernet networks commonly use a 6-byte MAC address.
2. Logical Address
- Logical Addresses are essential for universal communication, especially when data traverses different networks.
- Unlike physical addresses, which are local to specific networks and may be duplicated across multiple networks, logical addresses ensure seamless source-to-destination data delivery in an internetwork environment.
- Also known as an IP (Internet Protocol) Address, the Logical Address is employed at the network layer to universally identify devices like computers and routers.
-
IP addresses are globally unique, overcoming the limitations of local physical addresses. Currently, two versions of IP addresses are in use:
- IPv4: A 32-bit address format.
- IPv6: A 128-bit address format, designed to accommodate the growing number of connected devices on the Internet.
3. Port Address
- While a logical address ensures data transmission from source to destination, both the source and destination devices often host multiple processes engaged in communication.
- For instance, Users A & B may chat using Google Talk, while Users B & C exchange emails via Hotmail. Although the IP address facilitates data transmission from A to B, it doesn't specify the particular process on the destination device.
- To address this, a Port Address comes into play, providing a means to identify the source and destination processes. This ensures that data is not only delivered to the correct device but also to the correct process on that device.
- A Port Address serves as a name or label for a process and is a 16-bit identifier. For example, TELNET uses port address 23, and HTTP uses port address 80. Port addressing enhances the precision of data delivery in scenarios involving multiple processes on communicating devices.
4. Specific Address
- Port addresses facilitate data transmission from process to process, but challenges may arise in cases where multiple instances of the same process exist.
- For instance, consider Users A, B, and C chatting via Google Talk, each having two chat windows open. User A has two chat windows for B & C, and similarly, User B and User C have two chat windows each.
- While a port address ensures data delivery to the correct process (Google Talk) on User B, there are now two available windows for User A & C on B where the data can be delivered.
- To overcome this, specific addresses come into play. These user-friendly addresses help identify different instances of the same process. For example, multiple tabs or windows of a web browser operate under the same process (HTTP) but are identified using Uniform Resource Locators (URLs), serving as specific addresses to distinguish between various instances of the process.
