Topic 3 - Networks B Computer Science Notes

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 10

Networks

A computer network is a group of 2 or more computers and other hardware linked together
to enable communication and to share resources among a wide range of users with different
needs. This includes embedded devices and computing equipment that contains a processor,
such as network scanners and printers.

In order to establish a computer network 4 basic components are required:


- Communication devices
- Data transfer media
- Networking software
- Standards and procedures

Types of Network
They classify as different types according to functionality, scale, access and topology:
- Functionality: peer-to-peer, client to server systems
- Scale: WAN, LAN, VLAN, PAN
- Access: internet, intranet, extranet, VPN
- Network topologies:
- Bus: long line; terminators are in the end to bounce back signals.
- Ring: sends data in one direction only; really fast transport speeds as they do
not have anything stopping it.
- Star: direct communication between switch and computer.
- Mesh: all computers can communicate with each other; very resilient as you
do not lose any connection to other computers.

Wan:
- Much larger area
- Nodes connected by public networks
- Less secure
- Less transfer of data rates
- More expensive
- Large number of devices together
- Slower
- Hardware connectors

Client- server
- Terminals are connected to a network server, generally a powerful computer.
- Clients request data and services from servers.
- Servers store applications and data, and provide access to shared resources such as
printers and controls client access to these.
- Access is controlled through “user accounts” which have differing permissions based
on user needs and security requirements.
- Communication servers, e.g. web server, email servers, FTP servers, on the internet
function the same way.

Peer-to-peer
- All machines have equal status
- Each can communicate directly with all the others without the need for a server.
- Each acts as a client and a server, sharing resources.
- Software and data can be shared as long as the owner gives permission to the other.
- Often used for distributed processing systems
- Work has to be backed up at one of the machines.

P2P vs Client Server


P2P:
- Cheaper to set up
- Immune to central server failure
- Works best for smaller networks
- Security and backup are not centralised
- Access to resources depends on owner’s permission

Client-Server:
- Expensive to set up
- Cable to server damage can disable network
- Can manage large network
- Centralised security + backup
- Centralised controls on access to data

Standards
As different organisations use different OS and applications, communications between them
(over the internet) had to be standardised.

This requires the development of a protocol - a standard set of rules that enable devices to
communicate with each other.

They are important as if a programmer or hardware manufacturer were to use their own
invented protocol their programme would be unable to communicate with other programs.
The main internet protocol is called TCP/ IP - Transmission Control Protocol/ Internet
Protocol.
The OSI 7-layer model
- Application layer - closest to the user allows application to access network
resources. We interact with this layer.
- Presentation layer - transfer of data between formats. Data can be read.
- Session layer - creating a session between the two computers.
- Transport layer - prepares packets, establishing how much data to send. Uses TCP
(transmission control protocol).
- Network layer - adds the IP address of sender and destination IP address - router
level.
- Data link layer - responsible for transferring the data. Switch and NIC level.
- Physical layer - the physical media connecting devices.

All People Seem To Need Data Processing, layers can be


worked on independently.

VPN
If you are connected to a public network and you transfer data, your (unencrypted) data can
be intercepted by eavesdroppers.

A VPN allows a remote client to connect to a private network as if they were directly
connected to it. To the private network the remotely-connected client appears as just another
local node.

The remote user is able to access private LAN-based resources as if they were directly
connected.

This makes VPN an especially popular product for remote workers, particularly in fields
where security is important . The widespread use of VPNs have led to changes in working
patterns, such as workers telecommuting.

Transmission is encrypted, sender and receiver are both authenticated, IP addresses are
concealed, packets are hidden in other packets and encrypted (tunnelling).

Technologies needed to provide a VPN:

- VPN tunnelling software running on a server


- VPN aware router (and firewall)
- Encryption protocols (SSL/TSL)
- VPN client software (installed on the employees computer)
- A public network that at least the two nodes can connect to
- Both communicating nodes must be connected to the network, using VPN protocols.

VPN packets
Packets

- A unit of data, division of a larger file, made into a single unit that travels along a
given network path.
- Characteristics:
- Set amount of data
- Fixed structure
- Contains data sent via communications channel (payload)
- Specific details
- Packet switching:
- Broken into parts
- Sent independently via optimum route
- Reassembled at destination
- Each packet has IP address of sender and receiver

Packet

Header (12 bytes)

- IP sender
- IP receiver
- Protocol
- Sequence number

Data (112 bytes)

Actual data to be transmitted

Trailer/ footer (4 bytes)

- Transmission codes
- Error checking codes
- Control bits

Wireless networking

Advantages:
- Easy to set up & add new nodes - no cabling required
- Can rapidly be installed anywhere
- Allows a high degree of mobility compared to wired systems
- Have led to changes in working practices as a result
Disadvantages:
- Security - traffic can be easily intercepted
- Slow transmission speeds
- Interference is an issue caused by weather, other electronics etc.
- Signal range is affected by the user's own devices and antennae etc.
- Many standards are not compatible with one another.

Hardware required:
- WAP connected to a wired switch or router
- Wireless NIC in each wireless device → data link layer

Software:
- Wireless networking drivers
- Firmware in each NIC supporting shared networking standards → drivers have to prepare
data for the medium NIC is attached to.

Wireless networks VS Wired networks

- Reliability of transmission → wired is more reliable since it travels faster due to the fibre
optic cables that can travel a longer distance. Wireless is slower since it can suffer
interference, such as walls or glass or other radio signals.
- Wireless speed depends on the distance from the router, strength of the signal.
- Wired is immune to interference and is not affected by the surroundings.
- But the wired ethernet cable can be very easily broken and this could heavily affect
reliability.

- A wired network is more reliable than a wireless network in terms of transmission


because signal degradation is less of an issue as data travels through optic fibre cables
that can be transported through longer distances, on the other hand, wireless
network reliability signal degradation is more common as there are more
obstructions such as walls. Wired network is also faster as it has higher bandwidth
while wireless networks can offer high speed but in shorter distances and do not
reach the same speed as wired networks.

Characteristics

Wifi (‘wireless LAN’):


- Wireless Fidelity
- The most preferred home network type
- Latest wifi standards to allow backwards compatibility with older version
- Small transition radius (< 20m)

Worldwide Interoperability for Microwave Access (WiMAX):


- Designed for large distance high speed internet access
- Comparatively cheap method of provisioning internet over larger areas
- A rival to LTE (Long Term Evolution) standard
- Sometimes used as a wireless replacement for OSL (Over The Phone)

3G mobile:
- High-speed data internet access
- Cheap data transmission
- GPS
- SD video games
- 35 mile range, 300 kbps

Network security

Encryption types
- WEP (Wireless Equivalent Security) - obsolete. Superseded by WPA
- WPA (WiFi Protected Access) - 2004. Can be intruded from outside the network.
- WPAZ - 2006 onwards - intrusion only possible from outside the network.

UserIDs - Authentication
- Usernames & passwords
- 1, 2 & 3 - factor authentication:
- 1- something you know
- 2 - something you have
- 3 - something you are (biometrics)

MAC Addresses
- Assigned to networked devices during manufacture
- Used to identify pieces of hardware
- Managed by the IEEE (Institute of Electrical and Electronics Engineers)

Physical Security

- Locks on doors
- Attaching locks to hardware
Data Transmission

Protocol
A set of agreed rules between parties, to define the format and order of data transferred
between two devices. Actions to take during transmission. They ensure that data is
transferred correctly, without error, between two or more different electronic devices.

Confined to 1500 bytes. Both devices must be identified by IP addresses and ports.
Automatically detects and resolves missing data.

Example: TCP- enables two way communication between two devices.

Why do we need protocols?


- Data integrity: ensure data is not corrupted or altered during transmission.
- Flow control: ensure communication devices agree on transmission rates to reduce
effects of congestion.
- Congestion management: flow control helps networks avoid their capacity.
- Deadlock prevention: collision detection systems and time-to-live allow networks
to prevent situations where two devices are both waiting for one another to finish, so
never care.

Protocols- examples
- HTTP - Hypertext Transfer Protocol
- SSL - Secure Socket Layer: establishes encrypted connection between server and
client.
- IPSec - IP Secure: used for VPNs, authenticates and encrypts data packets.
- DHCP - Dynamic Host Configuration Protocol: issuing IP addresses to new nodes.

Why can speed vary across a network?

Dedicated LAN connections

UTP copper cable: 100 Mbps


Fibre optic cable: 50 Gbps
WiFi: 10 - 150 Mbps

Broadband (WAN) connections

DSL: 2 - 16 Mbps
Fibre optic: 20 - 100 Mbps
3G: approx. 1 Mbps
4G: approx. 20 Mbps
Factors affecting performance

- Bandwidth available
- Number of users (using network at the same time)
- (number of) data collisions
- Interference (walls, masonry, radio etc.)
- Distance data has to travel/ signal strength
- The transfer medium being used
- Amount of data being transferred
- Applications being used
- Server/ CPU performance
- Using a hub instead of a switch

Data compression

The process of reducing file sizes in order to improve the transmission speed or storage
requirements of a file. This is done by removing bits from the stored file. There are two kinds
of compression:

- Lossy
- Lossless

The more bits we can remove, the more data can be transmitted in the same time period as
would have been required to send a larger amount of data. The result of this is that data can
be disseminated more rapidly.

Lossy compression
- Involves removing data by taking an average of pixel colours over a wider sample
area.
- Most suitable for images, sound and video.

Lossless compression
- Does not lose any data at all because data is encoded, then completely restored
during the decoding process.
- Most suitable for text and some graphics

Transmission media
Description Speed Reliability Security Cost

Coaxial cable Much Very reliable but Reasonable; but Relatively cheap
slower can have some can be listened to
than optic interference
fibre

Fibre optic Fastest Very reliable but High - light does Expensive
can be not make a noise.
susceptible to Disruption is easily
breaks noticed

Wireless Depend on Can be blocked Can be intercepted More expensive


technology further easily and has relatively easily than cables for
connection limited range long distances

Twisted pair Faster Very reliable but Reasonable Relatively cheap


(UTP) than can have although they can
coaxial but interference be listened to
slower
than fibre
optic

Packet switching
Describes the type of network in which relatively small units of data called packets are routed
through a network based on the destination address contained within each packet. Its
predecessor was packet switching, which established a circuit between connected parties.

Advantages

- Packets are small and take different routes, resulting in better network utilisation
than circuit switching.
- Network resources are more fairly shared between network users as a result.
- Transmission errors can be detected rapidly.
- Packet switching results in packets taking different paths, including around
congestion or failed connections, therefore packet switched networks are resistant to
failure.

Disadvantages

- Bandwidth varies and cannot be guaranteed across the network.


- All participants must use the same network protocols.
- Packets don’t necessarily arrive in order.

You might also like