UNIT 1: Introduction to Computer Networks
Computer Engineering Section University Women’s Polytechnic Aligarh Muslim University, Aligarh
February 22, 2020
Outline
Computer Networks: What and Why??
Applications of Computer Networks Types of Network Connections
Categories of Networks: LAN, MAN & WAN Network Topologies
Network Configurations: Server Based, Peer-to-Peer & Hybrid Expanding Networks
Network Segmentation Types of Servers
Network Performance Parameters: Bandwidth, Throughput, Latency, Jitter etc.
Computer Network & Networking
A Computer Network is a set of autonomous comput- ers and other devices connected together to exchange infromation and sharing resources.
By autonomous we mean that no computer on the net- work can be forcibly start/stop by any other computer on the network.
Another definition is - A group of computers and other
devices connected together is called a network and the
concept of connected computers sharing information
and resources, is called Networking.
Centralized Computing Vs. Networking I
In centralized computing there is a computer with single powerful CPU and a number of terminals are connected to it as shown in the figure on slide 6.
All processing is performed within the boundaries of the same physical computer.
User terminals are typically dumb ones, incapable of functioning on their own. They are cabled to the central computer.
Sometimes these terminals have very little processing
power and memory of their own and are called intelli-
gent terminals.
Centralized Computing Vs. Networking II
This architecture places the tremendous burden on the central computer.
In recent years, there have been significant advances in the development of high performance personal comput- ers and networks.
There is now an identifiable trend in industry towards
downsizing, i.e. replacing expensive mainframe com-
puters with more cost effective networks of personal
computers that achieve the same or even better results.
Why Networks??
Networks are all about:
Sharing Resources.
Preserving Information.
Protecting Information.
Sharing Resources I
Resource sharing means sharing H/W and S/W.
Hardware Resource Sharing:
Networked Computers can share:
Printers, Fax Modem, Scanners, Hard Disks, Floppy Disks, CD-ROMS, Tape Backup Units, Plotters, Any device that can be attached to the Network.
Software Resource Sharing:
Software resources can be used more effectively over networks. With stand alone computers the software used on the computers must be present on each computer’s hard disk, whether or not that computer is used at that moment for the task the software performs.
For a large number of stand alone computers S/W cost can become more than expectations. It is also difficult and time consuming to install and configure the S/W individually on each computer.
Sharing Resources II
With a network we can install and configure the software on one com- puter (server) and can share it with other computers on the network.
We can also control the access to that S/W.
Not all S/W will use a network even if one is installed on the server. Dif- ferent S/W packages have different restrictions on how the software can legally be used on a network. We must check the S/W documentation to check what features the S/W provides in a networked environment.
Preserving Information
Besides, information and resource sharing a network al- lows information to be backed up to a central location.
Important information can be lost by mistake or acci- dentally when a stand alone computer has no backup means.
It is also difficult to maintain regular backups on a num- ber of stand alone computers.
In a networked environment when we take backup at a
central location from all computers, we have one place
to preserve it.
Protecting Information
With stand alone computer, access to the computer means access to the information on that computer.
Networks provide an additional layer of security by means of passwords.
We can give each network user a different account name
and password, allowing the network server to distinguish
among those who need access to have it and protecting
the information from tampering by those who do not.
Components of a Data Communication System I
There are f ive components of a data communication sys- tem:
1
Message: The message is the information (data) to be communicated. Popular forms of information include text, numbers, pictures, audio, and video.
2
Sender: A sender is a device that sends the data mes- sage. It can be a computer, workstation, telephone handset, video camera, and so on.
3
Receiver: A receiver is a device that receives the mes-
sage. It can be a computer, workstation, telephone
handset, television, and so on.
Components of a Data Communication System II
4
Transmission Medium: The transmission medium is the
physical path by which a message travels from sender to
receiver. Some examples of transmission media include
twisted-pair wire, coaxial cable, fiber-optic cable, and
radio waves
Components of a Data Communication System III
5
Protocol: A protocol is a set of rules that govern data
communications. It represents an agreement between
the communicating devices. Without a protocol, two
devices may be connected but not communicating, just
as a person speaking French cannot be understood by
a person who speaks only Japanese.
Transmission Modes I
The term Transmission Mode defines the direction of the flow of information between two communicating devices i.e.
it tells the direction of signal flow between the two devices.
There are three ways or modes of data transmission: Sim- plex, Half duplex (HDX) & Full duplex (FDX)
1 Simplex - unidirectional, one can transmit other can only receive, uses entire capacity of channel to send data in one direction. Ex- amples of simplex mode is loudspeaker, television broadcasting, television and remote, keyboard and monitor etc.
Transmission Modes II
2 Half Duplex - both station can transmit and receive but not at the same time, entire capacity of the channel can be used for each direction. Example of half duplex is a walkie- talkie in which message is sent one at a time and messages are sent in both the directions.
Transmission Modes III
3 Full Duplex - both stations can transmit and receive simultane- ously, either separate channels or channel can be divided between signals traveling in both directions.
Example of Full Duplex is a Telephone Network in which there is communication between two persons by a telephone line, through which both can talk and listen at the same time.
Classification of Computer Networks
There is no generally accepted taxonomy for computer net- works but two criteria based on which they can be classified are Transmission Technology and Scale.
Based on tramission technology they may be classified as:
Broadcast/ Multipoint Networks Point-to-Point Networks
Based on scale (physical limits) they can be classified as:
Local Area Network (LAN)
Metropolitan Area Network (MAN) Wide Area Network (WAN)
Broadcast Networks I
1 Broadcast networks have a single communication channel that is shared or used by all the machines on the network. Short messages called packets sent by any machine are received by all the others.
2 Broadcast systems generally use a special code in the address field for addressing a packet to all the concerned computers. This mode of operation is called broadcasting.
3 Some broadcast systems also support transmission to a subset of the machines known as multicasting.
4 Upon receiving a packet, a machine checks the address field. If the packet is addressed to it then the packet is processed, otherwise the packet is ignored.
5 The channel’s capacity is shared temporarily among the devices
Broadcast Networks II
Point-to-Point Networks I
1 A point-to-point connection is a direct link between two devices such as a computer and a printer. It uses dedicated link between the devices.
2 The entire capacity of the link is used for the transmission between those two devices.
3 In point to point networks, there exist many connections between individual pairs of machines.
4 To move from sources to destination, a packet (short message) may follow different routes.
5 In point-to-point connection, there can only be a single transmitter and a single receiver. On the other hand, in multipoint connection, there is a single transmitter, and there can be multiple receivers.
Point-to-Point Networks II
6 The switching nodes are not concerned with the contents of data.
Their purpose is to provide a switching facility that will move data from node to node until they reach the destination.
Classificaton Based on Scale I
Local Area Networks (LANs)
A Local Area Network (LAN) is a network that is restricted to smaller physical areas (few kilometers - typically upto 5 kms) e.g.
a local office, school, house, University campus etc.
On a ‘Local Area Network’ data transfer speeds are higher than WAN & MAN, and that can range from 10.0 Mbps (Ethernet network) to 10 Gbps (10 Gigabit Ethernet).
A LANs is usually implemented using any of the LAN Technolo- gies i.e. Ethernet, Token Ring and FDDI (Fiber Distributed Data Interface).
LAN Technologies are also referred to as Physical and Data Link Layers Protocols.
Classificaton Based on Scale II
Local Area Networks (LANs)
Each LAN technolgy offers specific devices (meant for that technol- ogy only) and network layouts (topologies) for connecting several computers as LAN.
The most prevalent LAN Technology isEthernet. It is standardized as IEEE 802.3.
The most common forms of Ethernet are 10BASE-T, 100BASE-T, and 1000BASE-T. All three use twisted pair cables. They run at 10 Mbit/s, 100 Mbit/s, and 1 Gbit/s, respectively.
Fiber optic variants of Ethernet are also very common in larger networks, offering high performance and longer distance (tens of kilometers with some versions).
Also referred to asWiFi, Wireless LANs (WLANs) are standardized as IEEE 802.11.
Classificaton Based on Scale III
Local Area Networks (LANs)
In general, network protocol stack software will work similarly on all varieties.
LANs are restricted in size. It simplifies network management.
They are more reliable as compared to MAN and WAN.
Metropolitan Area Network (MAN) I
A Metropolitan Area Network (MAN) is a network that connects two or more computers, communicating devices or networks in a single network that has geographic area larger than that covered by even a large ‘Local Area Network’ but smaller than the region covered by a ‘Wide Area Network’ (Typically from more than 5 km to 160 km).
A Metropolitan Area Networks bridges a number of ‘Local Area Networks’ with a fiber-optical links which act as a backbone, and provides services similar to what Internet Service Provider (ISP) provide to Wide Area Networks and the Internet.
MANs can bridge Local Area Networks without any cables by using microwave, radio wireless communication or infra-red laser which transmits data wirelessly.
Metropolitan Area Network (MAN) II
Distributed Queue Dual Bus (DQDB) is the Metropolitan Area Network (MAN) IEEE 802.6 standard for data communication.
Using DQDB, networks can extend up to 100km-160km and op- erate at speeds of 44 to 155Mbps.
Wide Area Network (WAN) I
Wide Area Network is a computer network that covers relatively larger geographical area such as a state, province or country.
It provides a solution to companies or organizations operating from distant geographical locations who want to communicate with each other for sharing and managing central data or for general com- munication.
WAN is made up of two or more Local Area Networks (LANs) or Metropolitan Area Networks (MANs) that are interconnected with each other.
In ‘Wide Area Network’, Computers are connected through public networks, such as the telephone systems, fiber-optic cables, and satellite links or leased lines. The ‘Internet’ is the largest WAN in the world.
Wide Area Network (WAN) II
WANs are mostly private and are build for a particular organization by ‘Internet Service Providers (ISPs)’ which connects the LAN of the organization to the internet.
Other Special Types of Networks I
1 Home Area Network(HAN): As the name would suggest, a HAN is the connection of network enabled devices in a domestic home.
2 Personal Area Network(PAN):
As the name suggests, a personal area network is intended for personal use within a range of a few ten meters.
Mainly, the technologies used for creating personal area networks are wireless.
Other Special Types of Networks II
A notable example of WPAN is the Bluetooth technology, mostly found on portable devices like smartphones, laptops, tablets, wearables, etc.
Other PAN technologies are Zigbee, Wireless USB, etc.
Wireless BAN can work together with PAN technologies.
For instance, a Bluetooth-connected smartphone can be used to sync data from wearables and various sensors present in the body.
3 Body Area Network(BAN):
You can create a body area network by using wearable devices like smart- watches, fitness bands, biometric RFID implants, and medical devices placed inside the body like pacemakers.
Wireless BAN is the primary form used to created such networks. It is defined as per the IEEE 802.15.6 standard which describes a short-range, extremely low power wireless communication within or in a vicinity of the human body.
Other Special Types of Networks III
4 Near-me Area Network(NAN):
A near-me area network (NAN) is a communication network that focuses on wireless communication among devices in close proximity.
Unlike local area networks (LANs), where the devices are in the same network segment and share the same broadcast domain.
The devices in a NAN can belong to different proprietary network infras- tructures (for example, different mobile carriers).
If two devices are geographically close, the communication path between them might, in fact, traverse a long distance, going from a LAN, through the Internet, and to another LAN.
Remember chatting with your friends on Facebook while all of you were sitting in the same room. You were part of a NAN, even though you might be on the networks of different carriers.
A message from your device would traverse all the way to Facebook servers over the internet come to your friend’s device sitting right next to you. In a logical way, both the device are on some sort of network.
Other Special Types of Networks IV
5 Storage Area Network(SAN):
SAN is a high-speed network of storage devices that also connects those storage devices with servers.
It provides block-level storage that can be accessed by the applications running on any networked servers.
Block-level storage is a type of storage in which each block (consisting of several hard drives) of the storage system can be controlled as an individual hard drive, and the blocks are managed by a server operating system.
SAN devices appear to servers as attached drives
Other Special Types of Networks V
6 Campus Area Network(CAN):
A network infrastructure covering the school, university, or a corporate premises can be dubbed as campus area network.
It can comprise of several LANs and connected to the internet using a leased line or any other means.
Other Special Types of Networks VI
7 Virtual Private Network(VPN):
A virtual private network (VPN) is a private network that is built over a public infrastructure.
VPN is a type of computer network which doesn’t have physical exis- tence.
The devices that are part of a VPN could be present anywhere on the earth, connected to each other over the internet.
Security mechanisms, such as encryption, allow VPN users to securely access a network from different locations via the Internet.
VPNs are used by corporates to interconnect their offices located in different places and give their remote employees access to company’s resources.
It has phased out another type of network known as Enterprise Private Network, a physical network created by organizations to link their office locations.
Network (LAN) Topologies I
A network topology is the arrangement of a network, including its nodes and connecting lines.
There are two ways of defining network geometry: the physical topology and the logical (or signal) topology.
On a network the way in which connections are made is called the physical topology of the network.
Physical topology specificaly refers to the physical layout of the network, especially the locations of the computers & devices and how cable is run between them (Geographical arrangement of com- puters, devices and cables.).
Logical (or signal) topology refers to the nature of the paths the signals follow from node to node.
Network (LAN) Topologies II
In many instances, the logical topology is the same as the physical topology.
But this is not always the case.
For example, some networks (e.g. Hub based star and Token Ring) are physically laid out in a star configuration, but they operate logically as bus or ring networks.
Physical Network (LAN) Topologies I
1 Bus Topology
On a bus network, the cable is just one or more wires, with no active electronics to amplify the signals or to pass them from one computer to another.
This makes the bus apassive topology.
When one computer sends a signal up (or down) the wire, all the other computers on the network receives the information.
But, only one (the one with the address that matches the address en- coded in the message) accepts the information. The rest discard the message.
Only one computer at a time can send the message.
Therefore, the number of computers attached to a bus can significantly affect the speed of the network.
A computer must wait until the bus is free before it can transmit.
These factors also affect star and ring networks.
Another important issue in bus network istermination.
Physical Network (LAN) Topologies II
As the bus is a passive topology, the electrical signal from a transmitting computer is free to travel the entire length of the cable.
Without termination, when the signal reaches at the end of the wire, it bounces back and travels back up the wire.
When a signal echoes back and forth along an unterminated bus, it is calledringing.
To stop signals from ringing we attach terminators at either end of the bus. Terminators absorb the signals and stop ringing.
Ethernet 10Base2 (uses BNC T-Connectors with coaxial cable), also referred to asthinnetis an inexpensive network based on bus topology.
Ethernet 10Base5also referred to asthicknet is another network based on bus topology. It uses thick coaxial cable compared to one used in thinnet.
Advantages
Simple, Reliable for small networks, Easy to use and install.
Physical Network (LAN) Topologies III
Scalable: Its easy to extend a bus.
Repeaters can be used to boost the signals if a larger bus is required.
Disadvantages
Heavy network traffic can slow down a bus considerably.
Problems caused by terminators.
Difficult to troubleshoot.
Physical Network (LAN) Topologies IV
Physical Network (LAN) Topologies V
2 Star Topology
Each computer on a star network communicates with a central device.
The central device resends the message to all computers (in a broad- cast star network(hub based)) or only to the destination computer (in a switched star network).
The central device in a star network can be anActive Hub or aPassive Hubor a Switch.
An active hub regenerates the electrical signal and sends it to all the computers connected to it (broadcast star network).
Active Hubs and Switches require electric power to run.
A passive hub such as wiring panel or punch-down blocks merely acts as a connection point and doesn’t amplify or regenerates the electrical signal.
Passive Hubs don’t require electric power to run.
We can useSTP(Shielded Twisted Pair), UTP (Unshilded Twisted Pair) orFiber Opticcables to implement star networks.
Physical Network (LAN) Topologies VI
Ahybrid hub/ switchcan accomodate several types of cable in the same star network.
Ethernet 10Base-T, 100Base-T (Fast Enternet), 1000Base-T (Gigabit Ethernet) and 10000Base-T (10 G Ethernet) are well known star net- works implemented with twisted pair cables (STP/UTP).
In above star based ethernets, if medium (twisted pair cable) is replaced by Fiber Optics then ’T’ gets replaced by ’F’ (e.g. 10Base-F etc.).
Advantages
It is easy to modify and add new computers to a star network without disturbing the rest of the network.
The centre of the star is a good place to diagnose network faults.
Single computer failure do not bring down the whole star network.
Several types of cable can be accomodated on same network.
Disadvantages
If central hub fails the whole network comes to halt.
Physical Network (LAN) Topologies VII
Active Hub and Switch consume electricity.
Physical Network (LAN) Topologies VIII
3 Ring Topology
On a ring network, each computer is connected to the next computer with the last one connected to the first.
Each computer retransmits what it receives from the previous computer.
The messages flow around the ring in one direction.
Since each computer retransmits what it receives, a ring is an active network and is not subject to the signal loss problem.
There is no termination because there is no end to the ring.
Some ring networks dotoken passing.
In token passing, a short message calledtokenis passed around the ring until a computer whishes to send information to another computer. That computer captures the token modifies it to adds an electronic address and data, and sends it around the ring.
Each computer in sequence receives the modified token and passes it to the next computer until either the electronic address matches the address
Physical Network (LAN) Topologies IX
The receiving computer sends an acknowledgement to the originator indicating that the message has been received.
The sending computer then creates a fresh token and place it on the network, allowing other computer to capture the token and begin trans- mitting.
This all happens very quickly. A token can circle a ring 200 meters in diameter at about 10,000 times a second.
Physical Network (LAN) Topologies X
The topology of IBM’s Token Ring LAN technology (standardized as IEEE 802.5) uses the similar kind of ring.
Physical topology of token ring is star but logically it is a token passing ring.
Some other ring networks have two counter-rotating rings, that help them recover from network faults.
The topology ofFDDI (Fiber Distributed Data Interface)LAN technol- ogy uses two counter-rotating rings.
Advantages
No computer can monopolize the network as each computer is given equal chance of capturing the token.
Token passing makes ring topology perform better than bus topology under heavy traffic.
Disadvantages
Failure of one computer on the ring can affect the whole network.
Physical Network (LAN) Topologies XI
Adding or removing computers disrupts the network.
4 Mesh Topology
In a mesh network topology, each of the network node, computer and other devices, are interconnected with one another.
In a full mesh topology, every computer in the network has a connection to each of the other computers in that network.
The number of connections in this network can be calculated using the formulan(n−1)/2, wherenindicates the number of devices.
Physical Network (LAN) Topologies XII
This type of topology is very expensive as there are many redundant connections, thus it is not mostly used in computer networks.
Advantages
Can handle high amounts of traffic, because multiple devices can trans- mit data simultaneously.
A failure of one device does not cause a break in the network or trans- mission of data (fault tolerant).
Adding additional devices does not disrupt data transmission between other devices.
Easy to troubleshoot.
Disadvantages
The cost to implement is higher than other network topologies, making it a less desirable option.
Building and maintaining the topology is difficult and time consuming.
Becomes unmanagable as more and more devices join network.
Hybrid Topologies I
In today’s networks we also see the combinations of the topologies of bus, star and ring.
A topology which is created by combining two or more topologies is referred to as a hybrid topol- ogy .
Some popular hybrid topologies are star-bus &
star-ring.
Figure on next page illustrates each of them.
Hybrid Topologies II
Picking a Right Topology (Ethernet LANs) You must look at following:
Cost.
Scalability / Extendibility.
Ease of installation.
Bandwidth Capacity.
Ease of fault isolation/tolerance/Maintenance.
Roles of a Computer on a Network I
A Computer can have one of the following three roles to play on a network:
Client: A Computer which uses but don’t provide Networking Resources.
They are also known as nodes or workstations (only when they run UNIX or its flavour)
Server: A computer which provides network resources.
Peer: A computer which provides as well as uses resources.
Can the role of a computer on a network be determined by looking at the operating system it is using?:
Clients run client operating system such as MS-DOS, OS/2 etc.
Peers run peer operating system such as Win 95/ Win 98/ Win NT workstation/ Win 2000 Professional/ Win XP (Professional and Home Editions)/ Win Vista/7/8/10 (Home & Professional Editions) etc.
Servers run Network Operating Systems such as Unix/ Linux/ HP-UX/
AIX/ Novell Netware/ Solaries/ Win NT/ Win 2000 Server/ Win 2000
Roles of a Computer on a Network II
Note that simply looking at the operating system we can’t decide whether a computer on a Network is a client or a peer or a server as almost all server operating systems can also be used as client OS.
So, a straight forward method to identify the role of a computer on a network is to look for its use on the network.
Based on the roles of the computers attached to them networks can also be classified as:
Server Based or Client/ Server Networks.
Peer-to-Peer Networks.
Hybrid Networks.
Server Based Networks I
Are identified by the presence of server(s) on the network.
Server provides security and administration of the network.
Client/ Server networks divide processing tasks between client and server.
Clients (referred to as front ends) request services such as file storage and printing, and servers (referred to as back ends) deliver them.
Server computers are typically more powerful than client comput- ers. Their hardware is optimized to function as servers.
Windows NT/ Server based networks are organized into domains.
A domain is a logical grouping of network computers that share a central directory database.
Server Based Networks II
A directory database contains user accounts and security informa- tion for the domain.
In a domain the directory resides on the computers that are con- figured as domain controllers.
A domain controller is a server that manages all security related user/ domain interactions and centralizes administration.
Advantages
Strong central security.
Centralized storage and backup.
Ability to share resources.
Optimized dedicated server which are faster than peers when sharing resources.
Server Based Networks III
Easy manageability of large number of users.
Disadvantages
Expensive dedicated servers.
Expensive Network Operating System and Client Licenses.
A dedicated network administrator.
Server Based Networks IV
Peer Networks I
There are no servers on a peer network.
Are defined by no central control over the network.
Users simply share resources.
Peer networks are organized into workgroups.
A workgroup is a logical grouping of networked computers that share resources (and strictly not the security information).
Each Windows (server or client flavour) computer in the workgroup maintains a local security database which contains a list of user accounts and resource security information for that computer.
As each computer in the workgroup maintains a local security database, the administration of user accounts and resource se- curity is decentralized.
Peer Networks II
A user must have a user account on each computer that the user needs to access.
Any changes to the user account information such as changing a password or adding new accounts must be made on each computer.
As already discussed, access to individual resources can be con- trolled if the user who shares the resource requires a password to access.
Since there is no central security trust users will have to know the individual passwords for each resource they wish to access. This can be quite inconvenient.
Peers are also not optimized to share resources.
When a number of users are accessing the resources on a peer, they notice significantly degraded performance.
Peer Networks III
Advantages
No extra investment on server.
Easy Setup.
No network administrator required.
Ability of users to control the share of resources.
Convenient for limited number of computers in close proximity.
Disadvantages
Additional load on computers because of resource sharing.
Inability of handling large networks.
Lack of central organization which can make data hard to find.
Requirement that users themselves administer the network.
Peer Networks IV
Hybrid Networks I
Hybrid computers have all types of computers operating on them and generally have active domains and workgroups.
It means that while most shared resources are located on servers, users still have access to any resources being shared by the peers in their workgroup.
It also means that network users don’t have to logon to the domain controller to access workgroup resources being shared by peers.
If users log on to the network with proper username and password they are eligible to enjoy “all shared” resources either they lie with the domain controllers or with the peers.
If users don’t log on the network with proper username and pass- word they are eligible to enjoy the resources contributed by the
Hybrid Networks II
Advantages
All advantages of Server based networks.
All advantages of Peer Networks.
Disadvantages
All disadvantages of Server Based and Peer Networks.
Hybrid Networks III
Expanding Networks: Principles & Devices
The objective is to expand a single network without breaking it into sub-networks.
Following devices can be used for this purpose:
Repeater.
Passive Hub.
Active Hub.
Switch.
Bridge.
Selection of device(s) strictly depends on a variety of factors.
Hubs and Switches have alredy been discussed.
So, we will examine rest of the devices.
Repeaters I
All transmission media attenuate (weaken) the electronic signal that travel through them.
Attenuation therefore limits the distance any medium can carry data.
Adding a device that amplifies the signal can allow it to travel farther, increasing the size of the network.
if we are connecting computers that are more than 100/ 500 me- tersapart using a10BASE-T/ 100BASE-T Ethernet cable, we will require a device that amlifies signal to ensure data transmission.
Devices that amplify signals in this way are called Repeaters.
Two types of repeaters are there -AmplifiersandSignal-regenerators (Boosters).
Repeaters II
Signal-regenerators (Boosters)create an exact duplicate of incom- ing data by identifying it amid the noise, reconstructing it and retransmitting only the desired information.
They reduce noise. The signal is boosted to its original strength and sent.
Because repeaters simply deal with the actual physical signals on a network, they operate at Layer 1 (Physical Layer) of ISO-OSI model.
Theoretically, repeaters can be used to combine an unlimited num- ber of cable segments.
Practically, network designs limit the number of repeaters.
An active hub / switch/router is by default a multiport repeater (signal-regenerator).
Repeaters III
Bridges I
Bridges connect network segments.
Unlike a repeater, which simply passes all signals it receives, a bridge selectively determinesthe appropriate segment to which it should pass a signal.
Bridges II
It does this by reading the source and destination MAC addresses of all the signals it receives.
For the network shown in figure, the process takes place as follows:-
The bridge receives all the signal from both segments A and B.
The bridge reads the destination address anddiscards (filters)all signals from segment A that are addressed to computers on segment A, because they do not need to crossover the bridge.
Signals from segment A addressed to a computer on segment B are retransmitted (regenerated)to segment B.
The signals from segment B are treated in the same way.
Bridges regenerate signals (physical layer activity) and read MAC addresses (data link layer addresses), so they are Layer 2 devices.
In future we will learn that a switch is actually a multiport-port bridge.
Assignment 2
Explain the following:
Bridge loop problem.
Spanning Tree Algorithm.
Network Segment
A network segment is a portion of a computer network that is separated from the rest of the network by a device such as a repeater, hub, bridge, switch, router or gateway.
Each segment can contain one or multiple computers or other hosts.
The nature of a segment depends on the nature of the device or devices used to interconnect end stations.
Accrdingly, there may be Layer 1, Layer 2 or Layer 3
segments in a network.
Collision & Broadcast Domains I
Collision Domain
A collision domain is a network segment connected by a shared medium or through repeaters where data packets may collide with one another while being sent.
A network collision occurs when more than one device attempts to send a packet on a network segment at the same time, leading to collision.
After a collosion has happened both devices must retransmit, one at a time. Not very efficient.
Members of a network portion connected by a hub or bus are in the same collision domain.
Collision & Broadcast Domains II
Broadcast Domain
A broadcast domain is the set of all devices on a network segment that hear all broadcasts sent on that segment.
When a host or server sends a network broadcast, every device on the network receives it.
For most of the hosts broadcast is of no use.
Too many broadcasts may result inBroadcast Stormswhich may flood the network.
Network Segmentation
The process of breaking up a larger network into a number of smaller ones is referred to as network segmentation.
Network segmentation is accomplished using bridges, switches, routers and gateways.
At some point of time we will have to break up one large network into a number of smaller ones.
This is because as the network grows LAN traffic also grows.
Possible causes of LAN traffic congestion are:
Too many hosts in a collision domain.
Too many hosts in a broadcast domain.
Broadcast Storms.
Low Bandwidth.
In following slides we shall study bridges, switches, routers and gateways from the perspective of network segmentation.
Bridge as Network Segmenter
Breaks the network into collision domains.
Clearly, it reduces network traffic and improves bandwidth.
Can’t break the broadcast domain(s).
Bridge doesn’t stop broadcast, instead forwards it in every port except the port on which it is received.
Makes forwarding decisions based on MAC addresses.
Improves existing network performance by breaking collision do- main(s), but can’t be used to create internetworks.
This is because they can’t filter the network based on Layer-3
Switch as Network Segmenter I
Like bridges, switches also break up collision domains.
Each port on a switch is a separate collision domain.
They are employed to add functionality to an existing LAN.
The main purpose of switches is to make a LAN work better i.e.
to optimize its performance.
Switches accomplish this by breaking up collision domain(s) and thereby providing more bandwidth for LAN users.
They ”switch” frames (based on MAC addresses) from one port to another within the switched network.
When a switch receives a broadcast it forwards the broadcast in every port except the port on which it is received.
Switch as Network Segmenter II
Theydon’tforward packets to other networks (i.e.) can’tfilter the network based on Layer-3 addresses (IP address).
So, they are not the devices for internetworking.
Bridge vs. Switch
Both bridge and switch break up collision domains on a LAN.
Looking at the way they do it, we conclude that a switch is just a multiport bridge.
Still somedifferences are there between switches and bridges.
Following table summarizes them:
Particulars Bridge Switch
No. of Ports Less number of ports (2-16). Large number of ports (100 or more)
Buffers No Buffers Buffer for each link
Error Checking Don’t Perform error checking Perform error checking
Cost Low Cost Comparatively high
Internetworking, Routable and Non-Routable Protocols I Before going into the details of other networking devices we need to discuss Internetworking, Routable and Non- Routable Protocols.
Internetworking is the process or technique of connect- ing different networks by using intermediary devices such as routers or gateways.
An internetwork needs protocols that allow it to identify each node on the network using following two things:
1 Address of the network.
2 Address of the node (device) itself.
Network protocols that provide both of these features
Internetworking, Routable and Non-Routable Protocols II Some common routable protocols are:
TCP/ IP(most popular today)
IPX/ SPX(Internetwork Packet Exchange/ Sequenced Packet Exchange (IPX/SPX) - Created by Novell for use on NetWare networks)
DECnet(created by Digital Equipment Corporation.)
AppleTalk(was a proprietary suite of networking protocols developed by Apple Inc. for their Macintosh computers.)
XNS(Xerox Network Systems (XNS) is a computer networking protocol suite developed by Xerox).
OSI (The Open Systems Interconnection protocols is a protocol suit developed jointly by the ISO and the ITU-T.Not to be confused with ISO-OSI model)
Internetworking, Routable and Non-Routable Protocols III A protocol that needs only a device address (such as MAC address) and not a network address is referred to as a non-routable protocol.
Some well known non-routable protocols are:
LAT(Local Area Transport Protocol from DEC).
NetBEUI(NetBios Extended User Interface from Microsoft).
Again, note that internetworking can only be realized
using routable protocols.
Router
Like bridges and switchs, routers also break up collision domains.
They don’t forward the network broadcast. When a router receives a broadcast it simply discards it.
So, routers are the only devices which can break the broadcast domains.
Routers route packets from one network to another, which means they can filter the network based on layer- 3 addresses (IP address).
Clearly, they are the devices which are used to create
internetworks.
Question???
How many collision and broadcast domains are there in the network shown below?
Answer
9 Collision Domains.
3 Broadcast Domains.
Brouter
A network device which combines the functions of a bridge and a router in one unite is called a brouter.
For non-routable protocols brouter operates at Layer 2 and is used as a bridge.
For routable protocols brouter operates at Layer 3 and is used as a router.
As networks continue to become more complex, a mix
of routable and non-routable protocols has led to the
need for a brouter.
Gateway I
Routers can successfully connect networks with similar protocols.
When the networks that must be connected, use com- pletely different protocols from each other, a more pow- erful and intelligent device is required.
A gateway is a device that can interpret and translate the different protocols that are used on two distinct networks.
This way a gateway is merely a multi-protocol router.
Gateways can be comprised of software, dedicated hard-
Gateway II
They operate at all seven layers of ISO-OSI model.
A gateway can actually convert data so that it works with an application on a computer on the other side of the gateway.
For example, a gateway can receive e-mail messages in one format and convert them into another format.
This shows that how a gateway works at all layers.
Therefore, we can connect networks with different pro-
tocols and architectures using a gateway.
OSI Model & Networking Devices
Types of Servers
Different servers do different jobs, from serving email and video to hosting Web sites.
Some commonly used servers are:
File Server Application Server Mail Server Web Server Print Server
Real Time Communication Server (Internet Relay Chat) Proxy Server
DNS Server Database Server
Multimedia or Media Server
File Server
On a network, a file server is a computer responsible for the central storage and management of data files.
Other computers on the same network can access these files.
A file server allows users to share information over a network without having to physically transfer files by pen drive or some other external storage device.
In its simplest form, a file server may be an ordinary PC that handles requests for files and sends them over the network.
In a more sophisticated network, a file server might be a dedicated network-attached storage (NAS) device.
It serves as aremote hard disk drive for other computers, allowing anyone on the network to store files on it as if to their own hard drive.
A program or mechanism that enables the required processes for file sharing can also be called a file server.
On the Internet, such programs often use the File Transfer Protocol (FTP).
Application Server
The application server is viewed as part of a three-tier appli- cation development, consisting of a graphical user interface (GUI) server, an application server, and a database and trans- action server.
More descriptively, it can be viewed as dividing an application into:
1 A first-tier,front-end, Web browser-based graphicaluser interface, usu- ally at a personal computer or workstation.
2 Amiddle-tier, application or set of applications, possibly on a local area network or intranet server.
3 A third-tier,back-end, database and transaction server, sometimes on a mainframe or large server.
In many usages, the application server combines or works with a Web server and is called a Web application server.
Mail Server
A mail server (sometimes also referred to an e-mail server) is a server that handles and delivers e-mail over a network, usually over the Internet.
A mail server can receive e-mails from client computers and deliver them to other mail servers.
A mail server can also deliver e-mails to client computers.
A client computer is normally the computer where you read/write your e-mails, for example your computer at home or in your office.
Also an advanced mobile phone or Smartphone, with e-mail capabilities, can be regarded as a client computer in these circumstances.
SMTP:Simple Mail Transfer Protocol POP:Post Office Protocol
Web Server
A web server is a program that uses HTTP (Hy- pertext Transfer Protocol) to serve the files that form Web pages to users, in response to their re- quests, which are forwarded by their computers’
HTTP clients (browsers).
The process is an example of the client/server model.
All computers that host Web sites must have Web server programs.
Leading Web server programs include Apache (the
most widely-installed Web server), Microsoft’s In-
Print Server
A print server, or printer server, is a device that connects printers to client computers over a network.
It accepts print jobs from the computers and sends the jobs to the appropriate printers, queuing the jobslocally.
A print server may be anetworked computerwith one or more shared printers.
Alternatively, a print server may be a dedicated device on the network, with connections to the LAN and one or more printers.
Print server functionality may beintegrated with other devices such as a wireless router.
A printer may have a built-in print server. Such printers are referred to asnetwork printers.
HP laserjet p2015dn printer in women’s polytechnic lab is a network printer.
Real Time Communication Server (IRC Server)
Internet Relay Chat (IRC) is an application layer protocol thatfacilitates communi- cation in the form of text.
The chat process works on a client/server networking model.
IRC clientsare computer programs that a user can install on his system.
These clients communicate with chat servers to transfer messages to other clients.
IRC is mainly designed for group communication in discussion forums.
But, it also allows one-on-one communication via private messages as well as chat and data transfer, including file sharing.
Proxy Server
In computer networks, a proxy server is a server (a computer system or an application) that acts as an intermediary for requests from clients seeking resources from other servers.
A client connects to the proxy server, requesting some service, such as a file, connection, web page, or other resource available from a different server.
The proxy server evaluates the request and fulfils it.
Today, most proxies are web proxies, facilitating ac- cess to content on the World Wide Web.
They provide anonymity and may be used to bypass
IP address blocking.
DNS Server
TheDomain Name System (DNS) is thesystem and network protocolused on the Internet.
It translates names of computers to numerical addresses ( IP addresses ) and vice versa.
It simply involves looking up in tables, in which names are linked to numbers.
DNS is a client-server system : a requestor (client) requests a name or address from a provider (DNS server ).
The server returns a reply.
Searching a number with a name is calledforward lookup.
Searching a name with a number is calledreverse lookup.
Database Server
A database server is a computer system that provides other computers with services related to accessing and retrieving data from a database.
Access to the database server may occur via a ”front end”
running locally at a user’s machine (e.g., phpMyAdmin), or
”back end”running on the database server itself, accessed by remote shell.
Multimedia or Media Server
A media server refers either to a dedicated computer appliance or to a specialized application software, ranging from an en- terprise class machine providing digital media on demand, to, more commonly, a small personal computer or NAS (Network Attached Storage) for the home, dedicated for storing various digital media (meaning digital videos/movies, audio/music,
Network Performance Measure I
Network Performance defines how good the network is.
Quality of Service (QOS) is an overall measurement of network performance.
From QOS point of view following measures are con- sidered important:
Bandwidth.
Throughput.
Latency.
Jitter.
Error Rate.
Network Performance Measure II Bandwidth
Bandwidth is defined as a range within a band of frequencies or wavelengths.
Bandwidth is also defined as theamount of data that can be transmitted through a medium in a fixed amount of time.
For digital devices the bandwidth is usually expressed in bits per second (bps) or less frequently in bytes per second.
For analog devices, the bandwidth is expressed in cycles per second or Hertz (Hz).
An increase in bandwidth in Hertz means an increase in the bandwidth in bits per second.
It is not generally possible to send more data than dictated by theShannon-Hartley Theorem (Shannon channel capacity
Network Performance Measure III Throughput
In general throughput means themaximum rateof production or the maximum rateat which something can be processed.
When used in the context of communication networks, such as Ethernet, throughput or network throughput is therate of successful message delivery over a communication channel.
Throughput is usually measured in bits per second (bit/s or bps)and sometimes in data packets per second (p/s or pps).
Latency
Network Latencyis the term used to indicate any kind ofdelay that happens in data communication over a network.
The speed of light imposes a minimum propagation time on all electromagnetic signals.
Network Performance Measure IV
It is not possible to reduce the latency below:
t =s/cm
where s is the distance and cm is the speed of light in the medium.
This approximately means 1 extra millisecond RTT (round- trip-time) for100km of distance between hosts.
Other delays also occur in intermediate nodes.
Network connections in which small delays occur are called low-latencynetworks.
Network connections which suffer from long delays are called
Network Performance Measure V Jitter
Jitter in an IP networks is the variation in the latency on a packet flow between two systems, when some packets take longer to travel from one system to the other.
Jitter results from network congestion and route changes.
Jitter is especially problematic in real-time communications like IP telephony and video conferencing.
Error Rate
In digital transmission, the number of bit errors (Ne) is the number of received bits of a data stream over a communica- tion channel thathave been altereddue to noise, interference, distortion etc.
Network Performance Measure VI
The bit error rate or bit error ratio (BER), is the number of bit errors (Ne), divided by thetotal number of transferred bits (N) during a studied time interval.
Mathematically, during the time interval ∆t,BER is expressed as follows:
BER = Ne N
BER is a unitless performance measure, often expressed as a percentage.
%BER = Ne
N ×100
