2. Reference Models and Network Devices – Express Learning: Data Communications and Computer Networks


Reference Models and Network Devices

1. What is meant by a reference model? Why do we need it?

Ans: A reference model is a conceptual layout that describes how communication between devices should occur. For efficient communication, the reference model identifies the tasks involved in inter-computer communication and divides them in logical groups called layers, with each layer performing a specific function. A communication system designed in such a manner is referred to as layered architecture. We need a reference model, as it provides various advantages, some of which are as follows:

It permits different types of network software and hardware to communicate with each other.
It defines standards for building network components thereby permitting multiple-vendor development.
It defines which functions should be performed at each layer of the model thereby promoting the standardization of network.
It divides the overall communication function into simpler and smaller components thus, helps in component development and reducing trouble shooting.

2. What are the advantages of layering in a network?

Ans: The main aim behind building a layered architecture is to reduce the design complexity of a computer network. Other advantages of layering in a computer network are as follows:

Layering can resolve complicated tasks by breaking it into smaller and manageable pieces.
Each layer can be analyzed and tested independently.
By layering, the functionalities are carried out in logical sequential manner.
Implementation of layers can be changed without disturbing other layers, as the details of all the layers are hidden from each other.
Layering allows reuse in a way that once a common functionality is implemented in a lower layer, the upper layers can share it.

3. Why standardization of network architecture is important?

Ans: Due to the advantages of layered architecture, many network vendors and suppliers used the concept of layered architecture for designing most computer systems, but the set of protocols and interfaces defined by each vendor were not same. In addition, the partition of layers defined by each vendor was alike. All this resulted into the integration incompatibility of different architectures defined by different vendors. Thus, to allow different vendor's network architectures to interoperate, the need for standardization of network architecture was felt. The standardization of architecture also tends to reduce the effort required to develop interfaces for the networking of different architectures.

4. What is Open Systems Interconnection (OSI) reference model? What are the principles used in defining the OSI layers?

Ans: OSI is a standard reference model for communication between end users in a network. By the term open system, we mean a set of protocols using which a system can communicate with any other system irrespective of the differences in their underlying hardware and software. In 1983, International Organization for Standardization (ISO) published a document called ‘The Basic Reference Model for Open Systems Interconnection’, which visualizes network protocols as a seven-layered model. The OSI model consists of seven separate but related layers, namely, physical, data link, network, transport, session, presentation and application layers as shown in Figure 2.1.

Figure 2.1 Layers in the OSI Model

With in a single machine, a layer in OSI model communicates with two other OSI layers. It services to the layer that is located directly above it while uses the services offered by the layer that is located directly below it. For example, the data link layer provides services to the network layer while calls upon the services of the physical layer. In contrast, during communication between two machines, each layer on source machine communicates with the corresponding layer (called the peer layer) on the destination machine using a set of protocols that are appropriate for the layer.

In 1983, Day and Zimmermann laid down certain principles which were used in defining the OSI layers. These principles are as follows:

Each layer should perform a well-defined function.
The functionality of each layer should be defined keeping in mind the internationally standardized protocols.
Changes made in any layer should not affect the other layers.
The numbers of layers should be more enough, so that each layer is associated with distinct functions.
A different layer should be made when the abstraction level changes.

5. Explain how data flows between the layers in OSI model.

Ans: When data is sent from one machine to another, it travels down sequentially from layer to layer on the source machine and as it reaches the destination machine, it moves up through the layers. While the data passes through layers on source machine, each layer adds header (and sometimes, trailer which is usually added at data link layer) to it and passes the whole unit to the layer directly below it. The header attached at each layer contains control information such as sequence numbers and size of data.

Figure 2.2 shows the data flow in OSI model during communication between two processes on machines A and B. Initially, the application layer (layer 7) of machine A adds an application header (AH) to the data and passes the package to the presentation layer (layer 6), which further adds its own header PH to the received data and passes it to the session layer. The same process is repeated at the session (layer 5), transport (layer 4) and network layers (layer 3). At the data link layer (layer 2), a trailer DT is also added to the data received from network layer along with the header DH. Finally, the entire package (data plus headers and trailer) reaches the physical layer where it is transformed into a form (electromagnetic signals) that can be transmitted to the machine B.

Figure 2.2 Data Flow in OSI Model

At machine B, the reverse process happens. The physical layer transforms the electromagnetic signals back into digital form. As the data travels up through the layers, each layer strips off the header (or trailer) added by its peer layer and passes the rest package to the layer directly above it. For example, the data link layer removes DH and DT from the package received from the physical layer and passes the resultant package to the network layer. The network layer then removes NH and passes the rest package to the transport layer and so on. Ultimately, the data reaching the application layer of machine B is in a format appropriate for the receiving process on machine B and thus is passed to the process.

6. Explain the duties of each layer in OSI model.

Ans: The seven layers of OSI model are divided into two groups according to their functionalities. Physical, data link and network layers are put in one group, as all these layers help to move data between devices. Transport, session, presentation and application layers are kept in other group, because they allow interoperability among different software. The functions of each layer are discussed as follows:

1. Physical Layer: This layer defines the physical and electrical characteristics of the network. It acts as a conduit between computers’ networking hardware and their networking software. It handles the transfer of bits (0s and 1s) from one computer to another. This is where the bits are actually converted into electrical signals that travel across the physical circuit. Physical layer communication media include various types of copper or fibre-optic cable, as well as many different wireless media.
2. Data Link Layer: This layer is responsible for reliable delivery of data from node to node and for providing services to the network layer. At sender's side, the data link layer divides the packets received from the network layer into manageable form known as frames. These data frames are then transmitted sequentially to the receiver. At the receiver's end, data link layer detects and corrects any errors in the transmitted data, which it gets from the physical layer. Other functions of data link layer are error control and flow control. Error control ensures that all frames have finally been delivered to the destination network layer and in the proper order. Flow control manages the sender to send frames according to the receiving capability of the recipient.
3. Network Layer: This layer provides end-to-end communication and is responsible for transporting traffic between devices that are not locally attached. Data in the network layer is called packet (group of bits) which in addition to data contains source and destination address. Packets are sent from node to node with the help of any of two approaches, namely, virtual circuit (connection-oriented) and datagram (connectionless). In virtual circuit method, route is decided while establishing connection between two users and the same path is followed for the transmission of all packets. In datagram method, there is no connection established; therefore, sequenced packets take different paths to reach destination. Therefore, virtual circuit method resembles telephone system and datagram method resembles postal system. Other functions of network layer include routing, deadlock prevention and congestion control. Network layer makes routing decisions with the help of routing algorithms to ensure the best route for packet from source to destination. Congestion control tries to reduce the traffic on the network, so that delay can be reduced and overall performance can be increased.
4. Transport Layer: The basic function of this layer is to handle error recognition and recovery of the data packets. It provides end-to-end communication between processes which are executing on different machines. It establishes, maintains and terminates communications between the sender process and the receiver process. It splits the message at the sender's end and passes each one onto the network layer. At the receiver's end, transport layer rebuilds packets into the original message, and to ensure that the packets arrive correctly, the receiving transport layer sends acknowledgements to the sender's end.
5. Session Layer: The session layer comes into play primarily at the beginning and end of transmission. At the beginning of the transmission, it lets the receiver know its intent to start transmission. At the end of the transmission, it determines whether or not the transmission was successful. This layer also manages errors that occur in the upper layers such as a shortage of memory or disk space necessary to complete an operation or printer errors. Some services provided by the session layer are dialog control, synchronization and token management. Dialog control service allows traffic to flow in both directions or in single direction at a time and also keeps track of whose turn it is to transmit data. Synchronization helps to insert checkpoints in data streams, so that if connection breaks during a long transmission then only the data which have not passed the checkpoint yet need to be retransmitted. Token management prevents two nodes to execute the same operation at the same time.
6. Presentation Layer: The function of this layer is to ensure that information sent from the application layer of one system would be readable by the application layer of another system. Therefore, presentation layer concerns with the syntax and semantics of the information transmitted. This is the place where application data is packed or unpacked and is made ready to use by the running application. This layer also manages security issues by providing services such as data encryption and compression, so that fewer bits need to be transferred on the network.
7. Application Layer: This layer is the entrance point that programs use to access the OSI model and utilize network resources. This layer represents the services that directly support applications. This OSI layer is closest to the end users. Application layer includes network software that directly serves the end users of the network by providing them user interface and application features such as electronic mail.

7. Explain the terms interfaces and services? Discuss protocol data unit (PDU).

Ans: Each layer contains some active elements called entities, such as process. Entities are named as peer entities when they are in the same layer on different systems. Between two adjacent layers is an interface which defines the operations and services of the lower layer that are available to its immediate upper layer. A well-defined interface in a layered network helps to minimize the amount of traffic passed between layers. The set of operations provided by a layer to the layer above it is called service. The service is not concerned about the implementations of operations but defines what operations the layer can perform for its users. Thus, lower layer implements services that can be used by its immediate upper layer with the help of an interface. The lower layer is called a service provider and the upper layer is called a service user.

A layer can request for the services of the lower layer, which is present below it, through a specific location known as service access point (SAP). SAP has a unique address associated with it. For example, in a fax machine system, SAP is the socket in which a fax machine can be plugged and SAP addresses are the fax numbers which are unique for every user. Therefore, to send a fax, the destination SAP address (fax number) must be known (Figure 2.3).

For communication and information sharing, each layer makes use of PDUs. PDU can be attached in front (header) or end (trailer) of the data and contains control information which is encapsulated with data at each layer. Depending on the information provided in the header, each PDU is given a specific name. For example, at transport layer, data plus PDU is called a segment; at network layer, segment and PDU (added by network layer) is given the name packet or datagram and at data link layer, packet with data link PDU is called a frame. The PDU information attached by a specific layer at the sender's end can only be read by the peer layer at the receiver's end. After reading the information, the peer layer strips off the PDU and passes the remaining package to its immediate upper layer.

Figure 2.3 Layers and Their PDU

8. Write the functions of data link layer in OSI model.

Ans: Data link layer is responsible for the transmission of frames between two nodes and provides error notification to ensure that data is delivered to the intended node. To achieve this, it performs the following functions:

Framing: The data link layer takes the raw stream of bits from the physical layer and divides it into manageable units called frames. To indicate the start and end of each frame to the receiver, several methods including character count, bit stuffing and byte stuffing are used.
Physical Addressing: The data link layer adds physical address of the sender and/or receiver by attaching header to each frame.
Flow Control: The data link layer provides flow control mechanism, which prohibits a slow receiver from being flooded by the fast sender. If the sender's transmission speed is faster as compared to the receiving capability of receiver, it is quite probable that some frames are lost. To avoid such undesirable events, the data link layer must provide a means to control the flow of data between sender and receiver.
Error Control: The data link layer is responsible for ensuring that all frames are finally delivered to the desired destination. To achieve this, the error control mechanism of data link layer makes the receiver to send positive or negative acknowledgement to the sender. Positive acknowledgement gives surety to the sender that frame has been received without any errors and negative acknowledgement indicates that frame has not been received or has been damaged or lost during transmission. It is the responsibilty of data link layer to ensure the retransmission of damaged and lost frames.
Access Control: When two or more devices are connected to each other via same link then data link layer protocol detects which device has control over the link at a point of time. The Institute of Electrical and Electronis Engineers (IEEE) has divided the data link layer into two sublayers.
  • Logical Link Control (LLC) Sublayer: This sublayer establishes and maintains links between the communicating devices. It also provides SAPs, so that hosts can transfer information from LLC to the network layer.
  • Media Access Control (MAC) Sublayer: This sublayer determines which device to access the channel next in case the channel is being shared by multiple devices. It communicates directly with the network interface card (NIC) of hosts. NIC has MAC address of 48 bits that is unique for each card.

9. Discuss some functions of the session layer and presentation layer.

Ans: Session layer is responsible for the establishing and maintaining session between processes running on different machines as well as synchronizing the communication between them. Some other functions of this layer are as follows:

It allows the processes to communicate in either half duplex or full duplex mode. That is, information can be transmitted between the processes either only in one direction at a time or in both directions at same time.
It makes use of checkpoints for synchronization that helps in identifying which data to retransmit.

The presentation layer is concerned with the syntax and semantics of the information being transmitted between communicating devices. Other functions of presentation layer are as follows:

It converts the representation of information used within the computer to network standard representation and vice versa.
It encrypts and decrypts data at the sender's and receiver's end respectively. It also compresses the data for reducing the traffic on the communication channel.

10. Compare the functionalities of network layer with transport layer.

Ans: Both network and transport layers are responsible for end-to-end communication but still there are certain differences in the set of services they provide. These differences are listed in Table 2.1.

Table 2.1 Comparison Between Network and Transport Layer

  Network layer   Transport layer
• It performs routing and makes the routing decisions based on the priority of packets. • It ensures that the entire message is delivered to destination machine's transport layer without any error.
• It is a connection-oriented layer. That is, it requires connection to be established between source and destination before delivering the packets. • It provides both connection-oriented and connectionless services.
• It performs packet sequencing, flow control and error control. • It provides end-to-end data transport service.

11. Explain in brief the transmission control protocol/Internet protocol (TCP/IP) reference model.

Ans: TCP/IP model was developed by the U.S. Department of Defense (DoD) to connect multiple networks and preserve data integrity. TCP/IP protocol model came after the OSI model and the numbers of layers in TCP/IP differ from that of the OSI model. TCP/IP model comprises of four layers, namely, network access (also called host-to-network layer), Internet, transport and application layers (see Figure 2.4).

Figure 2.4 TCP/IP Model

The network access layer of TCP/IP model corresponds to the combination of physical and data link layers of OSI model. The Internet layer corresponds to the network layer of OSI model and the application layer performs tasks of session, presentation and application layers of OSI model with the transport layer of TCP/IP performing a part of responsibilities of session layer of OSI model.

TCP/IP protocol suite contains a group of protocols forming a hierarchy such that the lower layer protocols support upper layer protocols.

1. Network Access Layer: This layer does not rely on specific protocol hence supports all standard protocols. It connects two nodes to the network with the help of some protocol and move data across two nodes which are connected via same link. The nodes after connection can transfer IP packets to each other. This layer is also referred to as host-to-network layer.
2. Internet Layer: The main function of this layer is to enable the hosts to transmit packets to different networks by taking any of the routes available for reaching the destination. This layer strengthens the whole architecture and defines the format of packet. The rules to be followed while delivering the packets are transparent to the users. Internet layer supports many protocols such as IP, address resolution protocol (ARP), reverse ARP (RARP), Internet control message protocol (ICMP) and Internet group message protocol (IGMP). IP is an unreliable and connectionless protocol that transmits data in the form of packets called datagrams. Each datagram is transmitted independently and can travel through a different route. Moreover, the datagrams may reach the destination not necessarily in the order in which they were sent or may be duplicated. IP neither keeps track of the routes followed by the datagrams nor does it perform any error checking. It tries its best to deliver the datagrams at their intended destinations; however, it does not ensure the delivery. ARP is used to identify the physical address of a node whose logical address is known. RARP performs just the reverse of ARP, that is, it enables a host whose physical address is known to identify its logical address. This protocol is used when a new node is connected to the network or when an already connected node is formatted. ICMP is used to send error messages to the sender in case the datagram does not reach its destination. IGMP is used to deliver the same message to a number of recipients at the same time.
3. Transport Layer: The main function of this layer is to deliver a message from a process on the source machine to a process on the destination machine. This layer is designed to allow end-to-end conversation between peer entities. It uses three protocols, namely, TCP, user datagram protocol (UDP) and stream control message protocol (SCMP) to accomplish its responsibilities. TCP is a connection-oriented protocol which means a connection must be established between the source and the destination before any transmission begins. It is also a reliable protocol, as it ensures error-free delivery of data to the destination. UDP is an unreliable and a connectionless protocol that performs very limited error checking. SCTP is the combination of UDP and TCP and it supports advanced features such as voice over the Internet.
4. Application Layer: This layer contains all the high-level protocols such as file transfer protocol (FTP) and virtual terminal (TELNET). Some more protocols which were added later include domain name service (DNS), hyper text transfer protocol (HTTP) and many more. With the help of various protocols, this layer integrates many activities and responsibilities for effective communication.

12. Describe various types of addresses associated with the layers of TCP/IP model.

Ans: Each layer in the TCP/IP model uses an address for the efficient delivery of data between communicating nodes. The host-to-network layer (physical plus data link layer) relates to physical address, network layer relates to logical address, transport layer concerns with port address and application layer defines specific address. The description of these addresses is as follows:

Physical Address: It is the address assigned to a node by the network (LAN or WAN) in which it is connected. It is the lowest-level address that is included in the frames at the data link layer to identify the destination node. The size and format of physical address is highly dependent on the underlying physical network. That is, different networks can have different address formats. Physical address is also known by other names including link address, MAC address and hardware address.
Logical Address: In an inter-networked environment connecting different networks having different address formats, the physical addresses are inadequate for communication. Thus, a universal addressing system is used that assigns each host in the network a unique address called logical address (also referred to as IP address or software address) which is independent of the underlying physical network. For example, in Internet, each host connected to the Internet is assigned a 32-bit IP address and no two hosts connected to the Internet can have the same IP address. It is the responsibility of the network layer to translate the logical addresses into physical addresses.
Port Address: The data link layer (using physical address) and network layer (using IP address) ensure end-to-end delivery of data, that is, data is reached the destination host. Now, since multiple processes may be running simultaneously on the host machine, there should be some means to identify the process to which data is to be communicated. To enable this, each running process on the host machine is assigned with a label what is known as port address. Using the port address, the transport layer ensures process-to-process delivery. In TCP/IP architecture, port address is of 16 bits.
Specific Address: Some applications such as e-mail and World Wide Web (WWW) provide user-friendly addresses designed for that specific address. Some examples of specific address include an e-mail address that helps to identify the recipient of that e-mail and URL of a website that helps to search a document on the web.

13. Compare OSI model with TCP/IP model.

Ans: OSI and TCP/IP are layered models that allow the computer systems to communicate with each other. The OSI reference model was developed by ISO in order to standardize the protocols being used in various layers and the TCP/IP model was developed by DoD to connect multiple networks. Both models have some similarities which are as follows:

Both OSI and TCP/IP models use set of independent protocols for enabling communication between users.
In both the models, upper layers focus on application such as web browser and lower layers focus on end-to-end delivery of data.

The differences between OSI and TCP/IP models are listed in Table 2.2.

Table 2.2 Differences Between OSI and TCP/IP Models

  OSI model   TCP/IP model
• It is a seven-layer model. • It is a four-layer model.
• It was unable to connect to radio and satellite network. • It had the ability to connect to radio and satellite network.
• It supports only connection-oriented communication in transport layer while both connection-oriented and connectionless communication in network layer. • It supports both connection-oriented and connectionless communication in transport layer while only connectionless communication in network layer.
• It clearly distinguishes services, interfaces and protocols. • It does not clearly distinguish services, interfaces and protocols.
• It was defined before the invention of the Internet. • It was defined after the invention of Internet.
• The model was developed before the corresponding protocols came into existence. • The model was developed after the protocols came into existence.

14. Explain where the following fit in the OSI reference model.

(a) A 4-kHz analog connection across the telephone network.

(b) A 33.6-kbps modem connection across the telephone network.

(c) A 64-kbps digital connection across the telephone network.

Ans: (a) The actual 4-kHz analog signal exists only in the physical layer of the OSI reference model.

(b) A 33.6-kbps modem is used for connecting a user to the switch across the telephone network and a modem also performs error checking, framing and flow control. Therefore, data link layer will be used for performing such functionality.

(c) A 64-kbps digital signal carries user information and is similar to the 4-kHz analog connection which makes use of twisted pair cable. Therefore, physical layer will be used for performing such functionality.

15. Discuss in brief the Novell Netware network model.

Ans: Novell Netware is the popular network system which was designed for replacing mainframes with a network of PCs thereby reducing the cost of companies. In this network, each user is assigned a desktop PC operating as client that uses services (database, file etc) provided by some powerful PCs operating as servers. Novell network is the modification of old Xerox network system (XNS) and it uses a protocol stack (see Figure 2.5). It was developed prior to OSI and looks like TCP/IP model.

Figure 2.5 Novell Netware Model

The physical and data link layers of Novell Netware network can use any standard protocols including Ethernet, IBM token ring and ARCnet. The network layer uses Internet packet exchange (IPX) protocol, which is an unreliable and connectionless inter-network protocol. IPX transmits packets from the source to destination transparently regardless of whether the source and destination are on the same or on the different networks. It provides the same functionality as that of IP. However, the only difference between the two is that IPX uses 12-byte address while IP uses 4-byte address. The transport layer uses network core protocol (NCP) which is a connection-oriented protocol. It provides many services in addition to transporting data. Another protocol which works on transport layer is sequenced packet exchange (SPX) that provides only data transport service. The application can choose from a variety of protocols such as SAP and file server.

16. Explain various network devices.

Ans: Networks are becoming more complicated and more pervasive everyday. Therefore, to reduce complication, some network devices were developed. Network devices help nodes to get connected in a network for efficient communication. Network devices include NIC, switch, router, bridge and gateway.

Network Interface Card

It is a hardware device that connects clients, servers and peripherals to the network through a port. Most network interfaces come as small circuit board that can be inserted onto one of the computer motherboard's slots. Alternatively, modern computers sometimes include the network interface as part of their main circuit boards (motherboards). Each network interface is associated with a unique address called its MAC address. The MAC address helps in sending information to the intended destination. NICs are the major factor in determining the speed and performance of a network. It is a good idea to use the fastest network card available for the type of workstation one is using (Figure 2.6).

Figure 2.6 Network Interface Card


It is the most basic device on a network. Signals that carry information within a network can travel a fixed distance before attenuation endangers the integrity of the data. A repeater installed on the link receives signal, regenerates it and sends the refreshed copy back to the link. Doing this means that the new signal is clean and free from any background noise introduced while travelling down the wire. In Figure 2.7, two sections in a network are connected by the repeater.

Figure 2.7 Repeater

Repeaters are most commonly used to extend a network. All network cable standards have maximum cable length specification. If the distance between two network devices is longer than this specification, a repeater is needed to regenerate the signal. Without the repeater, the signal will be too weak for the computers on each end to reliably understand. A good example of the use of repeaters would be in a LAN using a star topology with unshielded twisted pair cabling. The length limit for unshielded twisted pair cable is 100 m. The repeater amplifies all the signals that pass through it allowing for the total length of cable on the network to exceed the 100 m limit. Nonetheless, the repeaters have no in-built intelligence and they do not look at the contents of the packet while regenerating the signal. Thus, there is no processing overhead in sending a packet through a repeater. However, a repeater will repeat any errors in the original signal.


It is a small box that connects individual devices on a network, so that they can communicate with one another. The hub operates by gathering the signals from individual network devices, optionally amplifying the signals, and then sending them onto all other connected devices. Amplification of the signal ensures that devices on the network receive reliable information. A hub can be thought of as the centre of a bicycle wheel, where the spokes (individual computers) meet.

Nowadays, the terms repeater and hub are used synonymously, but actually they are not same. Although at its very basic level, a hub can be thought of as a multi-port repeater. Typically, hubs have anywhere from 4 to over 400 ports. When a signal is received on one port of the hub, it is regenerated out to all the other ports. It is most commonly used to connect multiple machines to the same LAN. Administrators connect a computer to each port on the hub, leaving one port free to connect to another hub or to a higher-level device such as a bridge or router.


This device allows the division of a large network into two or more smaller and efficient networks. It monitors the information traffic on both sides of the network, so that it can pass packets of information to the correct location. Most bridges can ‘listen’ to the network and automatically figure out the address of each computer on both sides of the bridge. A bridge examines each packet as it enters though one of the ports. It first looks at the MAC address of the sender and creates a mapping between the port and the sender's MAC address. It then looks at the address of the recipient, comparing the MAC address to the list of all learned MAC addresses. If the address is in the list, the bridge looks up the port number and forwards the packet to the port where it thinks the recipient is connected. If the recipient's MAC address is not in the list, the bridge then does a flood; it sends the signal to all the ports except the one from where it was received. As a result, a bridge reduces the amount of traffic on a LAN by dividing it into two segments. It inspects incoming traffic and decides whether to forward or discard it (Figure 2.8).

Figure 2.8 Bridge

Bridges can be used to connect networks with different types of cabling or physical topologies. They must, however, be used between networks employing the same protocol. Since a bridge examines the packet to record the sender and looks up the recipient, there is overhead in sending a packet through a bridge. On a modern bridge, this overhead is miniscule and does not affect network performance.


It is a multi-port bridge. It connects all the devices on a network, so that they can communicate with one another. The behaviour of a switch is same as that of a bridge. It is capable of inspecting the data packets as they are received, determining the source and destination device of that packet, and forwarding that packet appropriately. The difference is that most switches implement these functions in hardware using a dedicated processor. This makes them much faster than traditional software-based bridges.


It is an essential network device for interconnecting two or more networks. The router's sole aim is to trace the best route for information to travel. As network traffic changes during the day, routers can redirect information to take less congested routes. A router creates and/or maintains a table, called a routing table that stores the best routes to certain network destinations. While bridges know the addresses of all computers on each side of the network, routers know the addresses of computers, bridges and other routers on the network. Routers can even ‘listen’ to the entire network to determine which sections are the busiest. They can then redirect data around those sections until they clear up (Figure 2.9).

Figure 2.9 Router

Routers are generally expensive and difficult to configure and maintain. They are critical components of a network and if they fail, the network services will be significantly impaired. Most routers operate by examining incoming or outgoing signals for information at the network layer. In addition, they can permit or deny network communications with a particular network.


It is an internetworking device, which joins networks operating on different protocols together. It is also known as protocol converter. A gateway accepts the packet formatted for one protocol and converts the formatted packet into another protocol. For example, a gateway can receive e-mail message in one format and convert it into another format. A gateway can be implemented completely in software, hardware, or as a combination of both. One can connect systems with different protocols, languages and architecture using a gateway (Figure 2.10).

Figure 2.10 Gateway

17. What are the different types of bridges?

Ans: A bridge is a network device that allows division of a large network into two or more similar networks. It connects two segments of LANs and monitors the traffic on both sides of the network, so that it can pass packets of information to the correct location. It maintains a forwarding table that links addresses of all stations connected through it and, thus, helps to forward frames from one station to another. Different types of bridges are as follows.

Transparent Bridge

As the name implies, the existence of bridge is transparent to the stations connected through it. The transparent bridge is also called learning bridge, as its forwarding table is made automatically by learning the movement of frames in the network. Initially, the forwarding table contains no entries; however, as the frame move across the networks, the bridge uses the source addresses to make or update entries to the forwarding table and the destination address to make forwarding decisions. Figure 2.11 shows a transparent bridge connecting two networks LAN1 and LAN2 via ports 1 and 2, respectively.

Figure 2.11 Transparent Bridge

To understand how transparent bridge works, suppose station A wishes to send frame to station D. Since there is no entry corresponding to A or D in the forwarding table, the bridge broadcasts the frames to both the ports (that is, ports 1 and 2). However, at the same time, it learns from the source address A that the frame has come through port 1 and, thus, adds an entry (the source address A and port 1) to the forwarding table. Next time, whenever a frame from a station (say, C) destined for A comes to the bridge, it is forwarded only to port 1 and not elsewhere. In addition, an entry (source address C and port 2) is added to the forwarding table. Thus, as the frames are forwarded, the learning process of bridge continues.

An important advantage of transparent bridge is that since stations are not aware of the presence of bridge, the stations need not be reconfigured in case the bridge is added or deleted.

Source Routing Bridge

This bridge is used to connect two or more token ring LANs. In addition to the source and destination address, each frame includes the address of the all the bridges to be visited as specified by the source station. Thus, the path of a frame is already defined by the source station and intermediate nodes cannot take any decision on the routes. The source station acquires the addresses of bridges by exchanging some special frames with the destination station before the transmission of data frame begins. Figure 2.12 shows a source routing bridge.

Remote Bridge

It is used to connect two bridges at remote locations using dedicated links. Remote bridge configuration is shown in Figure 2.13 in which two bridges are interconnected using WAN.

Figure 2.12 Source Routing Bridge

Figure 2.13 Remote Bridge

18. Differentiate bridge, router and repeater.

Ans: Some differences among bridge, router and repeater are listed in Table 2.3.

Table 2.3 Comparison Among Bridges, Routers and Repeaters

  Bridge   Router   Repeater

• It operates at the data link layer of OSI model.

• It operates at the network layer of OSI model.

• It operates at the physical layer of OSI model.

• It is mostly used in LANs.

• It is mostly used in internetworking.

• It is used to extend the physical length of a network.

• It transmits the frames in similar network.

• It transmits the packets using similar protocols in different networks.

• It is an analog device connected to two cable segments which regenerates the signals.

19. Differentiate switch and hub.

Ans: The differences between switch and hub are listed in Table 2.4.

Table 2.4 Differences Between Switch and Hub

  Switch   Hub
A switch operates at the data link layer. A hub operates at the physical layer.
It is a complex device and more expensive than a hub. It is a simple device and cheaper than switch.
It is a full-duplex device and more secured. It is a half-duplex device and less secured.
Each port of switch has its own collision domain, that is, each port has buffer space for storing frame; therefore, when two frames arrive at the same time, frames will not be lost. The entire hub forms a single-collision domain, that is, when two frames arrive at the same time, they will always collide and hence frame will be lost.
It is an intelligent device, as it maintains a table to transmit the frame to the intended recipient. It is a non-intelligent device, as each time frame is broadcast to all the connected nodes.
It utilizes the bandwidth effectively. Wastage of bandwidth is more in the case of hub.

20. Differentiate router and switch.

Ans: Some of differences between the router and switch are listed in Table 2.5.

Table 2.5 Differences Between Router and Switch

  Router   Switch

• It connects nodes on different networks.

• It connects nodes within a same network.

• It operates in network layer.

• It operates in data link layer.

• It uses IP address for transmission of packets.

• It uses MAC address for transmission of frames.

• It is more intelligent and complex device than switch.

• It is less intelligent and simpler device than router.

• Various algorithms are used to forward packets along their best path.

• No such algorithms are used by switch.

• It needs to be configured before using.

• Most switches are ready to use and need not be configured.

Multiple Choice Questions

1. The correct order of corresponding OSI layers for having the functionalities of packet priortization, shared access resolution, end-to-end flow and error control and socket-based inter process communication are__________.
(a) network, physical, transport and application
(b) network, data link, presentation and application
(c) network, presentation, data link and transport
(d) network, data link, application, presentation
2. The main function of transport layer is__________.
(a) synchronization
(b) node-to-node delivery
(c) process-to-process delivery
(d) updating routing tables
3. IP is responsible for__________communication while TCP is resposible for__________communication.
(a) process-to-process, host-to-host
(b) host-to-host, process-to-process
(c) process-to-process, node-to-node
(d) node-to-node, host-to-host
4. Which of the following is/are an application layer service?
(a) File transfer and access
(b) Domain name service
(c) Remote login
(d) All of these
5. To interconnect two homogeneous WANs, we need a__________.
(a) bridge
(b) gateway
(c) router
(d) hub
6. A bridge recognizes the addresses of__________.
(a) physical layer
(b) data link layer
(c) network layer
(d) application layer
7. Which of the following internetworking device uses the greatest number of layers in the OSI model?
(a) Bridge
(b) Gateway
(c) Router
(d) None of these
8. In Novell Netware, IPX is a/an__________byte address.
(a) 8
(b) 12
(c) 4
(d) 48
9. Which type of bridge builds and updates its tables from address information of packets?
(a) Tranparent
(b) Source routing
(c) Remote
(d) None of these
10. Which of the following address(es) is needed for the effective communication?
(a) MAC
(b) IP
(c) Port address
(d) All of these


1. (b)

2. (c)

3. (b)

4. (d)

5. (c)

6. (b)

7. (b)

8. (b)

9. (a)

10. (d)