DISTRIBUTED NETWORKING - By Hansa Edirisinghe

DISTRIBUTED NETWORKING

BSc (Hons) Assignment 2011

DISTRIBUTED NETWORKING

Hansa K. Edirisinghe

BSc (Hons) University of Portsmouth, UK

3/1/2011

Case Study

After establishing a reputation in the Arts and Design Markets, the Singapore Arts Academy

(SAA) considers expanding and investing into a new Multimedia department and courses by

setting up campuses around Asia pacific region.

Currently, the Academy has a very poor data communication network. As you have designed

most of the other institutions’ network infrastructure, the dean of the Academy contracted you to

design a sound network that will cope with the current and the future demands.

Some facts about the current network at the Academy:

The school is located on the two-storey of the converted shopping mall in the City State of

Singapore. Each floor has several departments. Each department has 10 to 15 members of staff.

The ground floor comprehends the administration and the account departments. For payment,

the administration and the account departments are considering the use of smart cards based on

Biometrics technologies and Radio Frequency Identification (RFID). The Academy has a

corporate minicomputer (a UNIX computer) that processes all accounting data, and a Microsoft

File Server for students support.

The first floor accommodates a team of 30 academics, three IBM compatible labs, and two

multimedia labs. The two-multimedia labs are to be equipped with the latest multimedia tools

and technology. Each lab contains 20 personal computers, including wireless access points. The

applications running on two servers include Lotus Notes, remote access, student/staff workbench

and a management information system (MIS). Twenty of the existing classrooms in the school

are not connected to the network. The faculty, staff and students require communication with

each other via electronic mail, and have access to the Internet.

As the Academy expands, the network connections were put in as required. There has been some

discussion about establishing a corporate website as well. The dean is interested in running a

collaboration application like groupware, Voice over IP (VoIP) and Video Conferencing with

University partners in Boston, USA, and London, UK.

The current infrastructure runs on a token ring network with a capacity of 4Mbps. Furthermore,

the communication between various departments, connected to the network and the University

partners, is established via emails and Internet services.

Requirements of the proposed network for the Academy:

1) To implement a LAN and WLAN for the Academy which replace the current 4Mbps token

ring network,

2) To implement access points for wireless access for the students, staff and guests,

3) To link up all the office networks together on to a backbone network to support higher

volume of data,

4) To Identify and evaluate the state-of-art smart card Biometrics Technologies, and the Radio

Frequency Identification (RFID) mechanisms for payment,

5) To include further network expansion at the Academy to tolerate additional increases in the

network traffic such as the immediate inclusion of 20 computers, and to set up a web server

in order to establish a corporate website,

6) To implement a WAN to automate the data transfer between all the offices and the

University partners’ networks,

7) To provide a high bandwidth platform for collaboration application like groupware, Voice

over IP (VoIP) and Video Conferencing with the University partners and other campuses,

8) To identify and resolve key issues relating to network security for the school.

THERE ARE FIVE MARKS OVERALL ALLOCATED, FOR USE OF LANGUAGE

AND PRESENTATION

QUESTION 1

Discuss suitable LAN and WLAN technologies for the Academy. Critically evaluate your

proposed solution in terms of protocols, topology and cabling used. A details description of the

protocol, the advantages and pitfalls of the topology and a technical specification of the cables is

mandatory.

QUESTION 2

Critically discuss how to minimize the network congestion and the data access between

Academy campuses and the University partners’ networks using WAN connectivity. Your

discussions should include two different possible types of networks to connect between the two

continents. Your answer should also include the protocols associated, data rate, cost, and

performance aspect.

QUESTION 3

Prepare a comprehensive contingency plan for the Academy network in terms of strategy and

appropriate hardware/software solutions to ensure network availability. You should include four

significant threats that would affect network availability in case of major service disruptions and

propose counter measure mechanisms.

QUESTION 4

Critically evaluate three key strategic network security safeguards that can help to secure the

Academy Network. You should provide a detailed description of how these security mechanisms

can be used to secure the network and a comprehensive assessment of the advantages and

disadvantages of your proposal.

QUESTION 5

Design and explain, with the aid of a suitable fully labelled network diagram, how far the

proposed Academy Network meets the requirements listed. Provide a detailed assessment to

justify your proposed network, and any recommendations that you made for the specific

hardware and software technologies. You must clearly stated their usage in your proposal.

Page 1 of 23

Table of content

Question 1................................................................................................................................... 2

Question 2................................................................................................................................... 8

Question 3................................................................................................................................... 12

Question 4................................................................................................................................... 15

Question 5................................................................................................................................... 17

List of reference.......................................................................................................................... 19

Page 2 of 23

1. Question 1

Considering the current network of Singapore Arts Academy, We have to improve several arias

when establishing a strong data communication network within the branch campuses and among

the campuses.

Topologies which are going to use

Hardware the build the network

Reliability, effectiveness and cost

Contingency Plane

Network system Software

Token Ring is the currently using Technology which connects all the computers to ring or stare

and use a token passing methodology to transfer data or communicated between two computers.

When one computer needs to transfer data, it receives a free token, discard it and replace new

token with its own data, address to IP and release to the network. When the addressed IP

received data, it makes as received and passes it back to originating station, discard data frame

and release new token to the network. Token Ring data transfer rate is 4Mbps or 16Mbps. This is

a quite old technology and its devices are much more expensive. The main disadvantage of token

ring is only two stations in the network can send and receive data at one time.

Choose a LAN and WLAN technology

Considering about the requirement, I would like to propose 4Mbps Token Ring connection

replace with the Ethernet with the bandwidth of 1000Mbps as the Local Area Network of the

Academy. Ethernet use the Carrier Sense, Multiple Access / Collision Detect (CSMA/CD)

protocol to data communication.

CSMA/CD

Carrier Sense: Every nod is listening to the Ethernet wire

and check whether any device transmitting data. If

transaction happens, the nod waits while the current

transaction gets finish.

Multiple Access: It is possible to sensing (listening and

waiting to transmit) data from one or more devices at the

same time.

Collision Detection: If more than one ethane nods transmit

data the same time, Ethernet is able to detect collision.

Advantages of the Ethernet

An any network nod can send data in any time

Bandwidth availability 10, 100, 1000Mbps

Much easy to plane and low installation cost

Network components such as cables, Network interface cards and Connectors as low in

cost

Figure 1.1 CSMA/CD

Source : http://www.payer.de/cmc/cmc0801.gif

Page 3 of 23

Ethernet can work with Star topology, as well as Bus topology

Trustworthy and reliable when working with home and office network environments

It is easy to add and remove computers on the network

Supported by most Software and hardware systems

Disadvantages of Ethernet

A large traffic rate such as 40% - 50% may delay the whole network

Packet swishing produces visible delay

Which protocols to select

A protocol is the special set of rules that end points in a telecommunication connection use when

they communicate. (SearchNetworking.com 2011)

OSI and TCP/IP are the two models which

gather network protocols in to layers.

TCP/IP is an open source protocol which uses

to any kind of network, but there are some

restrictions also in the TCP/IP. In the SAA

network there are lot of changes can happen in

that protocol. So in side this protocol there is

advantages and disadvantages in the protocol

which use this university network. There are

IBM main frames compares also in this

Academy building. But the TCP/IP protocol is

unable to use in the IBM servers. So it will be

a pitfall in the university network in

Singapore. However as its mention the Art and

Design University the TCP/IP is a very reliable

protocol because less cost effect and it’s easy

to implement the network in any second.

Because the there are selected web sites or host or IPs has to block in the university. TCP/IP can

make change the routing configuration easy and make block the site and access the sites the user

need and wants. This makes advantage of the TCP/IP.

Selection of a Network Topology

Topology is a layout which devices can interact with each other. There are four major types of

topologies and there are plus points, as well as minuses in each and every topology. According to

them we should select the best suitable topology for our environment.

Topology Advantages Disadvantages

Bus Easy to control.

Easy to implement.

Most suitable for small networks

with 10 – 15 nods.

Cable length limited.

Number of stations limited.

Cannot perform well.

Page 4 of 23

Ring Transition data packets among two

nodes pass through the all other

nodes in the network.

Central server not required for

configuration.

Failure of single device or cable will down

the entire network.

Add and remove nod is difficult.

Star Simplicity of operation.

Add and remove nod is easy.

Failure of one node will not affect

to the whole network.

Error detection is easy.

Failure central device fails the entire

network.

Mesh No network traffic occurs.

Modification of a cable or nod does

not affect to the entire network.

High cost.

Considering of SAA environment, I would like to suggest Star Topology.

Star topology connects all the devices in the view of a

visual star. There is a central HUB or Switch. All the

other Nods (Computers, servers, printers, fax

machines, clock Machines) connected to the Central

Device. This central HUB or Switch acts as a repeater

for the data flow. The total data communication

network handles by the central device. Each and every

bit of data drives through the central device to its

destination nod. So a failure of one cable or failure of

one nod does not harm to the entire network. But

failures of the central device (it can be HUB or a

Switch) come out with an error the entire network

goes down. The nature (layout) of star network makes

really easy of trouble shooting within the network.

For the configuration of star, it is possible to use twisted

pair (STP or UTP) cables or Fiber-optic Cables.

Widely using cable type for Star is UTP cables

(Unshielded Twisted Pair cables) and the network port that is connected to both ends are RJ45

ports.

Looking at the Singapore Art Academy’s environment, it’s a two storied building. It’s facilitated

with 3 IBM and 2 Multimedia labs with two departments for Staff. Two departments and the File

servers are in ground floor.

Scenario:

An IBM lab equipped with 20 IBM PCs per each lab and Multimedia labs with 20 PCs.

According to the standard switch sizes, we have to use 24 port switches and connect the entire

lab PCs directly to switch as per the star topology layout. It was connected by UTP cables

Page 5 of 23

Figure 1.4 Hierarchical Star

Source : http://tjliu.myweb.hinet.net/COA_CH_12.htm

Hierarchical Star

The hierarchical topology is used to make

more connection ports available within an

building, office, university or Institute.

Plugging outlying switches (access switches)

into a master switch (Distribution or Core

switch) gives more hosts a place to plug in

without having to pull additional cable into

the area.

Advantages of Star

Star topology has simply

understandable layout

It makes easy to install the topology

Star performance is good. The way of

star transaction process, it performs

better than other network topologies.

When there is a communication goes on, just only tree network devices and two network

links actively participate to the communication process from the entire star. It can avoid

the unnecessary traffic within the network.

The rate of Error detection in star network devices, higher than other networks, because

of the simplicity of network structure.

If a failure occur in one nod does not harm the entire network connection and it is easy to

repair or replace the defected nod without breaking the entire network.

According to star structure, it is very easy to adding new devices and grove the network.

Disadvantages of Star

The entire network is highly depending on the central hub. If the hub (Switch) comes up

with error, the functionality of entire network goes down.

There are limited in the central hub. So there is a limit to grove the network and

connection of one new nod to star effect entire performance cycle of the network.

The performance level is dropped, when switch hands its high traffic, then the entire

network will get slow.

Network Cables

Within a wired network, cables play one of the main roles on the network.

Twisted-pair (Shielded and Unshielded twisted are the two categories of Twisted pair)

cables

Coaxial cable (very famously used before twisted pair)

Fiber-optic cables are the generally using network cable types. By looking at the cost of

cables, data transfer speed, usability and durability, we need to identify, what is the most

suitable cable type for our network.

Page 6 of 23

Twisted-pair cables

Two insulated copper wires warped around each other are Twisted-pair. When the electric

signals walking through the copper wire, it generate a small magnetic field around the wire

(according to Fleming’s right-hand thumb rule). The twisted nature of twisted pair cables is

minimize this disturbance. Ethernet over twisted-pair is not new within a building amount

several floors and several partitions.

Two typed of twisted-pair cables are;

Unshielded twisted pair (UTP)

Shielded twisted pair. (STP)

Unshielded twisted-pair (UTP)

UTP cable composed with eight copper wires which are covered by material which reduce

electromagnetic interferences. It reduces its electromagnetic interference according to the way it

twisted. UTP is low in cost and it is possible to use major networking architectures. Cable is very

flexible and easy to install. I suggest UTP cabling is the most suitably to an environment like

SAA. UTP cable can transfer from 10 Mbps to 1000Mbps in speed and the maximum cable

length for gigabit is Ethernet is 100 meters. Connecter would be RJ 45 and the cost is low. So

UTP is the most suitable cable type for SAA network. Because we have more than 120

computers and servers to connect with WAN and we selected 1000Mbps (gigabit) Ethernet to

implement the new network.

Shielded twisted-pair (STP)

STP has four pairs of copper wires and each pair is wrapped with a metallic foil, then these four

pairs again wrapped with a metallic foil. This is why we call it as Shielded and this reduces the

outer electromagnetic interferences and the inner electronic disturbances as well. Comparing

with the UTP cable STP prevents inner and outer disturbances better than UTP. But STP is

expensive and installation is difficult. Considering the SAA environment, it is not good enough

to go for STP cabling network according to the academy’s requirement and the STP cabling cost

probably be and additional cost to the academy.

Selection Cables for the topology

When we selecting a cable, we should the always look into bellow characteristics of a cable. The

characteristics which are

Maximum segmentation Length (max length of the cable that can transfer data)

Speed (speed in bits per second)

Cost (cheap or expensive)

Advantages and disadvantages of the using the selected cable type of cable.

Page 7 of 23

As above mentioned, my suggestion is UTP cable the most suitable cable for SAA networks out

of the other 4 cable types, according to requirements.

UTP Cable Standards

UTP cables divide into categories according to UTP standards. They are Cat 3 UTP, Cat 5 UTP,

Cat 5e UTP, Cat 6 UTP Cat 6A UTP and Cat 7 UTP. But we have to forget about the Cat 3 UTP

and Cat 5 UTP, because those can carry data only up to 10Mbps and 100Mbpd Ethernet.

Cat 6A UTP and Cat 7 UTP are designed for 10 Gbps Ethernet. So we are not necessary to go to

that much unwanted expense. Cat 5e UTP and Cat 6 UTP is only suite for Gigabit Ethernet.

Thus Cat 5e UTP can handle Gigabit Ethernet; I select Cat 6 UTP to configure the Academy

network.

Cat 6 UTP has backward compatibility. It support to the occurring rate of Crosstalk and noise

damage to the signal, keep on low level. Provide performance frequency up to 250MHz. Cat 6

UTP 10BASE-T UTP, 100BASE-TX (Fast Ethernet) UTP, 1000BASE-T/1000BASE-TX

(Gigabit Ethernet) UTP. Proper integrated Cat 6 UTP cables supports up to 10Gbps speed.

Page 8 of 23

2. Question 2

Page 9 of 23

Network Congestion

When a nod or device transmits more data at ones and overload, network congestion takes place.

This congestion reduces the accuracy and trustworthiness of network and communication data.

Because it delay the communication speed and the congestion might loess or block data packets.

When the switching nod getting congested and the round trip time duration increases, the

network problems are not addressable.

Going through the SAA network, we can find several points which can occur network

congestion.

When multiple links feed its data into to single link. It usually happen the points which internal

LANs are connected to WAN, such as the academy internal network links University partners in

Boston and London.

According to the proposed network, the academy network usage is very high. Other than the

current use wireless router implementation Smart card Biometrics Technology, RFID, newly

including computers, web server, VoIP Video Conferencing and new security futures require

more bandwidth and speed. So the routers in core network are capable to handle a large traffic.

Otherwise it will be another reason for occurring congestion.

Mainly the academy internal and external communication purposes and data and file transaction

purposes with partner Universities, TCP/IP such as Internet is essential. But it is a major point

which occurs of terrible network congestion, because the basic connection is less nature.

Minimise the network Congestion

We can use several telecommunication technologies to minimise congestions and sometimes it

makes fast data transmission. These telecommunication services designed for connect LAN and

WAN connectivity. As we discussed earlier, network congestion directly related theses services.

Frame Relay, X.25 and Asynchronous Transfer Mode (ATM) are some of them.

Page 10 of 23

Figure 2.1 Frame Relay Structure

Source : http://www.protocols.com/pbook/frame.htm

Frame Relay

Frame relay is a digital dedicated low cost data transmission service, which has higher data rate

intermittent traffic between Local area networks (LAN to LAN), between end points in Wide

area networks (WAN), remote accesses and access to internet. The information divided into data

packet, labelled with an ID and carries to the identified destination with very efficient way. This

process is known as packet switching technology. It sends data frames (data packets) over the

network for communication and checking errors while the transmission process. When there is

any error detection it retransmits the data.

In SAA network we can use Frame relay with the transition from Router. It makes our data

transaction fast and efficient.

This network service transmits data between remote networks with leased line access. Frame

relay has 56Kbps to 45Mbps range of bandwidth and it is possible to select the bandwidth which

is suit to our Academy usage. The physical connection to our frame relay circuit is known as port

speed.

Packet switching technology is also use with TCP/IP and X.25.

Frame Relay Structure

Protocol use in Frame Relay

The protocol which is using with frame relay is HDLC. HDLC stands for High-level Data Link

Control. HDLC is the protocol which manages the flow of data in Frame Relay. This is the most

common protocol using in layer 2 in OSI model.

ATM

Asynchronous Transfer Mode is a technology based on transfer data in cells or packets for a

fixed size. A cell which is used with ATM is small in size and it allows transmitting video, audio

and computer data along network.

So it is possible to use ATM to our technology between switches and end user computers of

multimedia Labs.

ATM creates a fixed cannel among two points when the beginning of the transmission process.

ATM utilized for high bandwidths like 34Mbps to upwards.

Page 11 of 23

X.25

X.25 is the advances version of Frame Relay. X.25 divides into three categories. Those are data

DTE, DCE and PSE. DTE stands for terminal equipment, DCT stands for data circuit-

terminating equipment and PSE stands for packet-switching exchange. DTE devices are

endpoints such as PCs, terminators, printers, scanners that communicate across X.25 networks.

DCE devices are communication devisers such as modems and packet switches. DCE are the

interface between DTE and PSE. PSE are swatches to compose bulk carriers of network.

Page 12 of 23

Figure 2.2 Characteristics between Frame relay and X.25

Source : http://www.dcbnet.com/notes/framerly.html

Figure 3.1 Switch

Source : http://lh4.ggpht.com/_9R2czlR5ELs/SoDac_wAD1I/AAAAAAAAA

Qs/5rl4f0gu5lc/img-xavb1004-back.jpg

3. Question 3

According to the structure of the Academy and considering about the new requirements there are

several hardware devices to configure a better network and software to control the network and

track the transaction. We have to consider about the several facts including reliability, efficiency

and cost.

Hardware

Switch

Switch is used to interconnect several

hosts(nods) together. There are switches in

different sizes. The common sizes are 4

port, 8 port, 16port, 24 ports and 48 port

switches and the size of the switch can be

selected according to the according to our

network. For each port there is a light

which blinks when it connected to

network. Switch operates at the data link

layer, which is the second layer of the OSI

model.

The MAC addresses of all the devices

which connect to switch are in its storage.

When a data frame passes through the switch, it examines the destination MAC address.

According to the Destination device MAC and the data frame’s destination MAC, switch decides

Page 13 of 23

Figure 3.2 Wireless Router Source : http://kb.netgear.com/app/answers/detail/a_id/1000/~/what-

is-a-router%3F

to which port, the data frame to copy. If the data frame’s destination MAC is unknown, then

switch will broadcast the frame to all of its ports.

We have 5 computer labs and two departments at the ground floor. There are several serves to

connect to the network. Because of that, this device is important to us. Make star network within

the IBM and Multimedia labs, connect servers including file, application and web, to the

Academy’s LAN. Switch can support 10/100/1000 Mbps port transfer rate and multiple switches

in the same network can support for different speeds. The price of a switch differs from brand to

brand, sizes and transfer rates. However a 24 port switch which supports for 1Gbps speed, we

need to allocate $100 - $150 from our SAA networking budget.

Wired Router and Wireless Router

Router is a device which transfer data to nods and work as a filter of data and its task is to copy

frames from one to another. It is a device which uses to connect on internet and other outside

networks. Router configuration allows you to apply filters, block accesses and deny access to

selected hosts. According to OSI model router operates at the network which is layer number

three.

Router use one more than one routing protocols

to create its routing table and looking at the

routing table router makes its decisions. Wired

routes are accessible with interface types, such

as DSL, Ethernet, Fast Ethernet, Fiber

Distributed Data Interface (FDDI) and Token

ring.

In the SAA network we need a wired router

and as well as wireless router. The academy’s

IBM and Multimedia labs have computers with

wireless access points and they require

implementing access points for wireless access for

the students, staff and guests.

Wireless routers are available with interfaces such

as IEEE 802.11(Wi-Fi)

How router works

Page 14 of 23

Threats affect to the network availability

Flood, Earthquakes, Tsunami

These are natural threats which we cannot stop or get rid of it. But we can minimise the damage

from these threats. For an example we have out academy database server and the backup server

in the same building. But will destroy all our property when sudden Tsunami disaster. If we keep

another backup server in our branch campuses in another country, it will minimise the damage

us.

Fire

Using fire alarm system and CO2 fire extinguishers will minimise the damage.

Sudden power failures

Sudden power failures can affect to computer software, hardware and network devices. UPS is a

device which supplies power from its battery for about 5 - 10 minutes, until power supply

connected to generators. Otherwise these UPS power supply time is enough to turn of the

devices.

Trojan horse, Viruses, Worm

Page 15 of 23

Figure 4.1 Attack Sophistication vs. Intruder Technical Knowledge

Source : www.andrew.cmu.edu/course/95-752/notes/netsec.ppt

These are small software which designed to delete or corrupt data in the computers, databases

and servers. This makes the PCs and servers to slow, and sometimes it harms to the network.

Using a fire wall and properly updated antivirus software can avoid these Virus infections.

Hackers and Unauthorized Access

People can come to our academy network without permission and harm or seal machined data

and network data. Not only the data, they can seal our hardware items as well. Using a password

protection to every PC is the primary protection for the network. Using different passwords even

from lab to lab is better for the security. Keep network administration passwords and server

passwords confidentially is essential. When the students, parents or staffs logging in to the

system there should be audit log. Audit log means there will be records of logins and the record

of the activities.

4. Question 4

It is essential to maintain a very good security system for large network like academy network,

which spread worldwide. There are a lot of ways which can attack the network.

Attack Sophistication vs. Intruder Technical Knowledge

Firewall

Firewall is a secure machine, which is works between a private network and a public network.

Normally firewall closely working with a router program, and check each and every network

packet to determine whether to forward it toward its destination.

Page 16 of 23

How Firewall Works…

There are lots of firewall screening methods. A Simple one is to screen requests to make sure

they come from acceptable domain name and Internet protocol address. Firewall use one of these

methods to control the traffic flowing in and out of the network.

Packet Filtering

Proxy service.

Shameful inspection

Biometrics authentication Techniques

Biometric authentication used to identify the person’s physical, biological characteristics

uniquely. This can be used in ongoing and absolute identification.

Eg: finger prints, Foot prints, Walking Style, DNA patterns, Voice

VPN

VPN is a private network which is used to

communicate with remote sites.

Page 17 of 23

5. Question 5

Branch campuses network

Page 18 of 23

Full network, main campus network and branch

Page 19 of 23

Reference

SearchNetworking.com (2011) accessed 4 March 2011

http://searchnetworking.techtarget.com/definition/protocol

Bibliography

Topologies Available at wikipedia.org (2011) accessed 18 Feb 2011

http://en.wikipedia.org/wiki/Network_topology

Advantages and Disadvantages of Topologies Available at buzzle.com (2011) accessed 27 Feb

2011 http://www.buzzle.com/articles/advantages-and-disadvantages-of-different-network-

topologies.html

Star Topology Available at exforsys.com (2011) accessed 21 Feb 2011

http://www.exforsys.com/tutorials/networking/star-topology.html

Advantages and Disadvantages of Tree Available at buzzle.com (2011) accessed 21 Feb 2011

http://www.buzzle.com/articles/star-topology.html and wikipedia.org (2011) accessed 21 Feb

2011 http://en.wikipedia.org/wiki/Star_network

Network Cables Available at freeonlineresearchpapers.com (2011) accessed 21 Feb 2011

http://www.freeonlineresearchpapers.com/lan-topology-disgn-cabling-specifications

UTP Cable Standards Available at tech-faq.com (2011) accessed 20 Feb 2011 http://www.tech-

faq.com/utp.html and comptechdoc.org (2011) accessed 21 Feb 2011

http://www.comptechdoc.org/independent/networking/guide/nettopology.html

Network Congestion Available at linktionary.com (2011) accessed 27 Feb 2011

http://www.linktionary.com/c/congestion.html

TCP Congestion Control Available at jpkc.ncwu.edu.cn (2011) accessed 27 Feb 2011

http://jpkc.ncwu.edu.cn/jsjwl/net/ross/book/transport_layer/congestion.html

Frame Relay Available at shopforbandwidth.com (2011) accessed 1 March 2011

http://www.shopforbandwidth.com/frame-relay.php

X.25 Available at networking.ringofsaturn.com (2011) accessed 2 March 2011

http://networking.ringofsaturn.com/Protocols/x25.php

Network Hardware Available at book.opensourceproject.org.cn (2011) accessed 3 March 2011

http://book.opensourceproject.org.cn/distrib/ubuntu/unleashed/opensource/0672329093/ch18lev1

sec4.html and tech-faq.com (2011) accessed 3 March 2011 http://www.tech-faq.com/types-of-

network-hardware.html