Why protocol required

Similarly, two hosts implementing the same protocol can connect and communicate easily with each other. Hence, protocols provide a common language for network devices participating in data communication.

Protocols are developed by industry-wide organizations. Support for network protocols can be built into the software, hardware, or both. All network end-users rely on network protocols for connectivity. There are different layers for instance, data, network, transport, and application layer, etc. It is the communication between entities in different systems, where entities can be a user application program, file transfer package, DBMS, etc.

Hence protocols can be implemented at the hardware, software, and application levels. A standard protocol is a mandated protocol for all devices. It supports multiple devices and acts as a standard. Standard protocols are not vendor-specific i. Proprietary protocols are developed by an individual organization for their specific devices. We have to take permission from the organization if we want to use their protocols. It is not a standard protocol and it supports only specific devices.

We may have to pay for these protocols. The key elements of the protocol determine what to be communicated, how it is communicated, and when it is communicated. Syntax refers to the structure or format of data and signal levels. It indicates how to read the data in the form of bits or fields. It also decides the order in which the data is presented to the receiver. Ad hoc networks establish a connection between two devices without an internet connection.

Passive optical networks PONs bring high broadband speeds and fiber to end users' doorsteps. IT pros should know what a PON is and how it can provide network solutions. Communication protocols allow different network devices to communicate with each other. They are used in both analog and digital communications and can be used for important processes, ranging from transferring files between devices to accessing the internet. Network management protocols define and describe the various procedures needed to effectively operate a computer network.

These protocols affect various devices on a single network — including computers, routers and servers — to ensure each one, and the network as a whole, perform optimally. Security protocols, also called cryptographic protocols, work to ensure that the network and the data sent over it are protected from unauthorized users. Network protocols do not simply define how devices and processes work; they define how devices and processes work together.

Without these predetermined conventions and rules, the internet would lack the necessary infrastructure it needs to be functional and useable. Figure 1. An analogy for the operation would be ordering some items online for delivery to your home.

You will be able to track the progress of the packages while they are in transit, and you may be required to sign for them, which provides acknowledgement of their delivery.

You will also be able to contact the seller of the items if any are not delivered to arrange for re-delivery. The four basic functions of TCP are: Ensuring data segments are delivered in the correct sequence to the correct application layer protocol.

Providing flow control, so segments are delivered at a rate the receiving device can handle. Multiplexing of multiple user applications, allowing them to simultaneously access the transmission network.

Error checking received segments, and requesting retransmission if segments have been corrupted. TCP breaks up data received from the application layer into small pieces known as segments. To provide reliable transmission, the segments are numbered before being passed to the IP process, which encapsulates them into packets.

TCP tracks the number of segments that are sent to a specific destination device from a specific application layer protocol. If it does not receive acknowledgement within a certain period of time, TCP assumes that the segments have been lost and will retransmit them. TCP also checks each segment to ensure that the contents have not been changed during transmission across the network media. This process is referred to as error checking , and is possible because each segment header includes a check-sum, which is a mathematical signature generated by feeding the data in the segment through a cyclical redundancy algorithm.

This is placed in the TCP header by the sender, and the receiving device will carry out the same calculation on the data in the received segment. If the signatures match, TCP will consider the data within the segment as not damaged. If the signatures do not match, TCP will arrange for the data to be retransmitted. TCP can serve multiple application layer protocols simultaneously, processing their data into segments and feeding them to the Internet layer in a process called multiplexing.

To allow TCP to deliver received segments to the correct application layer protocol, port numbers are used. Because TCP may receive a significant number of segments, which need to be error-checked, sequenced and delivered to the correct application layer protocol, it needs to be able to control the amount of segments it receives to allow it to operate effectively.

A TCP receiving process will agree a window size with a TCP sender, which dictates the amounts of segments that can be sent before a TCP acknowledgment is sent by the receiver. The TCP receiver can thus control the amount of segments it is sent, a process called flow control. Figure 2. A simplified windowing process is illustrated in the figure below. In a real TCP exchange, both the client and server would send data and windowing information, but for clarity only the server window size is shown.

TCP windowing and flow control process View larger image. Figure 4. The diagram below shows the same client and server exchange with a window size of However, the second segment, sequence number , is lost during transmission and is not received by the server: View larger image.

Figure 5. This is important for real-time programs, such as voice and video services, which work best with minimal delay between communicating devices. UDP does not use sequence numbers or windowing, so there is no need for a three-way handshake to set initial values. If a device using UDP becomes swamped by an excessive number of datagrams, it will simply drop those that it cannot process.

Because UDP does not use sequence numbers, it is unable to re-order datagrams that it receives in the wrong order. Figure 6. Figure 7.