COMPUTER NETWORKS

A computer network is a network of computers that are geographically distributed, but connected in a manner to enable meaningful transmission and exchange of data among them. Sharing of information, resources (both hardware and software), and processing load is the main objective of computer network.

Network topologies
Topology of a network refers to the way in which the networks nodes (computers or other devices that need to communicate) are linked together. It determines the various data paths available between any pair of nodes in the network. Although number of possible network topologies is seemingly limiters, four major ones are Star Network, Ring Network, Completely connected Network, and Multi Access Bus Network. They are described below. Choice of a topology for a computer network depend on a combination of factors such as : 1. 2. 3. 4. 5. 6. 7. Desired performance of the system. Desired reliability of the system. Size (number of needs and their geographical distribution) of the system. Expandability of the system. Cost of components and services required to implement the network. Availability of communication links. Delays involved in routing information from one need to another.

Star network
It has multiple nodes connected to a host node. That is, the nodes in the network are linked to each other through the host node and can communicate only via the

host node. The routing function is performed by the host node that centrally controls communication between any two other nodes by establishing a logical path between them.

Advantages
1. Star topology has minimum line cost because only one lines are required for connecting. 2. Transmission delays between two nodes do not increase by adding new nodes to network, because any two nodes are connected via two links only. 3. If any node other than the host node fails, remaining nodes are unaffected.

Host Node

A Star Configuration of computer network

Disadvantage
The system crucially depends on host node. If it fails, than entire network fails.

Ring Network
It shows a circular or ring network, in which each nodes has two communicating subordinates (adjacent nodes with which it can communicate directly), but there is no master node. A node receives data from one of its two adjacent nodes. The only decision a node has to take is whether the data is for its own use. If it is addressed to it, it utilizes it. Otherwise, it morally passes it to the next node.

Advantages
Ring network works well where there is no central node for making routing decisions. It is more reliable than a star network because communication is not dependent on a single central node. If a link, between any two nodes fails, or if one of the nodes fails, alternate routing is possible.

A Ring Configuration of computer network

Disadvantage
It is a ring network; communication delay is directly proportional to number of nodes in the network. Hence addition of new nodes in a network increases communication delay. It requires more complicated control software than star network.

Completely Connected network

It has a separate physical link for connected each node to any other node. Hence, each node has a direct link, called point to point link, with all other nodes in the network. The control is distributed with each node deciding its communication priorities.

A Completely Connected Configuration of computer network

Advantages

It is very reliable because any link failure will affect only direct communication between the nodes connected by that link.

Each node of the network need not have individual routing capability. Communication is very fast between any two nodes.

Disadvantages
It is the most expensive network from the point of view of link cost. If there are n nodes in a network, (n-1)/2 links are required. Hence, cost of linking the system grows with square of the number of nodes.

Multi Access Bus Network

In which all nodes share a single transmission medium. That is all nodes are attached to the same communication link (channel). When anode wants to send a message to another node, it appends destination address to the message and checks whether communication line is free. As soon as line becomes free, it broadcasts (places) the message on the line. As the message travel on the line, each node checks whether the message is addressed to it. The message is picked up by addressee node that sends an acknowledgement to source node and frees the line. This type of network is also known as Multipoint, Broadcast Network. It is appropriate for use in a local area network where a high speed communication channel is used and computers are confined to a small area. It is

also appropriate when satellite communication is used, because many computers at different geographical locations share one satellite channel.

Computer (Nodes)

Single Communication Line Shared By All Nodes

A Multi Access Bus network

Advantages
It helps in reducing the number of physical lines. Failure of a node does not affect communication among other nodes in the network. Addition of new nodes to the network is easy.

Disadvantages
All nodes in a network must have good communication and decision making capability. If the shared communication line fails, entire networks fail.

Hybrid Network
Different network topologies have their own advantages and limitations. Hence, in reality, a pure star or ring or completely connected network is rarely used. Instead an organization normally uses a hybrid network that is a combination of two or more different network topologies. Exact configuration of a network depends on needs and structure of the organization. In some cases hybrid network may have components of star, ring, and completely connected networks.

Ring

Star

Completely Connected

A hybrid network having a combination of ring, star, and completely connected network topologies.

Types of computer Networks

Networks are classified broadly into two types Local Area Network (LAN) Wide Area Network (WAN) WAN are also referred to as long haul networks key characteristics often used to differentiate between these two types of networks are : 1.

Geographic distribution :
A LAN is restricted to a limited geographic coverage of a few kilometers, but a WAN may extend over several thousand kilometers. Therefore, LAN typically provide communication facilities within a building or a campus, whereas WAN may operate nationwide or even worldwide.

2. Data Rate
Data transmission rates are usually much higher in LAN than in WAN transmission rates in LAN usually range from 10 megabit per second (Mbps) to 1 Gigabit per second (Gbps). On the other hand, transmission rates in WAN usually range from 1200 bps to 2 Mbps. The data transmission rates are increasing continuously for both LAN and WAN with advancements in technology.

3. Error Rate
LAN generally experience fewer data transmission errors than WAN do typically, bit error rates are in the range of with LAN as opposed to with WAN.

4. Communication Link
The most common communication links used in LAN are twisted pair, cable, and fiber optics on the other hand typical communication links used in WAN are telephone links, microwave links, and satellite channels because the nodes are physically distributed over a large geographical area.

5. Ownership
A LAN is owned typically by a single organization because of its limited geographic coverage. A WAN, however, is formed usually by interconnecting multiple LAN each of which may belong to a different organization.

6. Communication cost
The cost of transmitting data in a LAN is negligible, since transmission medium is owned usually by user organization. However, with a WAN, this cost may be very high because transmission media used are leased lines or public communication systems such as telephone lines, microwave links, and satellite channels.

Communication Protocols
A protocol is a set of formal operating rules, procedures, or conventions that govern a given process. A communication or network protocol, therefore, describes rules that govern transmission of data over communication networks. These rules provide a method for orderly and efficient exchange of data between sender and receiver and for proper interpretation of controls and data transmitted as raw bits and bytes. These rules are embedded in data communication software.

Role of a Communication Protocols

In any computer network, a Communication Protocols normally performs following functions for efficient and error free transmission of data. It has a separate set of rules (implemented in software) for performing each of these functions.

1. Data sequencing
It refers to breaking a long message into smaller packets of fixed size. data sequencing rules define method of numbering (or sequencing) packets to detect loss or duplication of packets, and to identify packets belonging to the same message.

2. Data routing
Routing algorithms are designed to find the most efficient path between source and destination nodes of a message. They can handle varying degree of traffic on a given network configuration with optimal time utilization.

3. Data formatting
Data formatting rules define which group of bits or characters within a packet constitutes data, control, addressing, or other information.

4. Flow control
A communication protocol also prevents a fast sender from flooding a slow receiver with data. It ensures resources sharing and protection against traffic congestion by regulating flow of data on communication

5. Error control
These rules are designed to detect errors in messages and to ensure transmission of correct messages. The most common method for

correcting errors is to retransmit an erroneous message block. This method requires coordination between sender and receiver nodes, so that the black having error is discarded by receiver node and is retransmitted by sender node.

6. Precedence and order of transmission

These rules ensure that all nodes get a chance to use communication lines and other resources of a network based an priorities assigned to them.

7. Connection establishment and termination

These rules define how connections are established, maintained, nad terminated when two nodes of a network want to communicate with each other.

8. Data security
Providing data securing and privacy is also built into most communication software packages. It prevents access of data by unauthorized users.

9. Log information
Several communication software are designed to develop log information consisting of all jobs and data communications tasks that have taken place. Such information may be used for charging users of a network based on their usage of network resources.

Concept of layered protocols in network design

Modern computer networks are designed in a modular fashion for easy and electric signals, light signals or electromagnetic signals, depending on whether the two nodes are on a cable circuit respectively. Even electrical details, such as how many volts to use 0 to 1, how many bits can be sent per second, and whether transmission can take place only in one direction or in both directions simultaneously, are decided by physical layer protocols. Physical layer protocols also deal with mechanical details such as size and shape of connecting plugs number of pins in plugs, and function of each pin. RS232-C is a popular physical layer standard for serial communication lines.

Data link layer

Physical layer simply transmits data from senders node to receivers node as raw bits. Data link layer is responsible for detecting and correcting any error in transmitted data. Since physical layer is concerned only with a raw bit stream, data link layer partitions it into from, so that error detection and correction can be performed.

The OSI Model

Initial computer networks had their own set of standards and conventions that were hardware dependent. Each manufacture used to develop its own communication protocols is defined for each layer in this seven layer architecture. Hence, each layer has an independent function and deals with one or more specific aspects of communication. In actual implementation of these seven layers, first three layers are likely to be in hardware, next two in operating system, presentation layer in library subroutines in users address space, and application layer in users program.

Physical layer
Physical layer is responsible for transmitting raw bit stream between two nodes it converts sequence of binary digits into efficient handling of the system. They are normally split into a series of modules and are logically compared of a succession of layer. Each layer offers certain services to higher layers, shielding those layers from details of how the offered services are actually implemented.

Network Interface Cards

Network interface card, often referred to as NIC or network card is a hardware device that allows a computer to be connected to a network, both functionally and physically. The NIC is a printed circuit board that is installed on to one of the expansion stets of a computer, and provides a port to which a network cable is attached. It is an add on card that is connected directly to a computers I/O bus.

Network Layer
Network layer is responsible for setting up a logical path between two nodes for communication to take place it encapsulates frames into packets that can be transmitted from are node to another by using a high level addressing and routing scheme. Routing is the primary job of network layer; two popular network layer protocols are the X.25 and internet protocol (IP).

Transport layer
Transport layer accepts messages of arbitrary length from session layer, segments them into packets, submits them to network layer for transmission, and finally resembles packets at destination. Transport

layer protocols include mechanisms for handling lost and out of sequence packets. The transport layer assigns a sequence number to each packet and user sequence numbers for detecting lost packets and for ensuring that messages are reconstructed in correct sequence. The two most popular TLP are TCP and UDP.

Session layer
Session layer provides means of establishing, maintaining, and terminating a dialogue or a session between two end users. It allows communicating parties to authenticate each other before establishing a dialog session between them it specifies dialog type one way, two way alternatives, or two ways simultaneous and initiates a dialog session, if message is a connection request message. It also provides priority management service that is useful for giving priority to important and time bound message over normal, less important message.

Presentation layer
Presentation layer provides facilities to covert message data into a form that is meaningful to communicating application layer entities. It may perform such transformations as encoding and decoding, code conversion, compression and decompression encryption and decryption, on message data depending on applications requirements.

Application Layer
Application layer provides services that directly support end users of network. It is a collection of miscellaneous protocols for various commonly used applications such as email, file transfer etc. some popular protocols are X.400 (email protocol), X.500 (directory server protocol), FTP.

Submitted By : Mohd Amil 13-AEM-21