J276/01 - Section 2 - Networks
LANs
WANs
Topolgies
Protocols
Packet Switching
Stands for Local Area Network and covers a small geographical area loacted on a single site. All the hardware for a LAN is owned by the organisation that uses it.
LANs are either wired with Ethernet cables or wireless using Wi-Fi. LANs are often found in businesses, schools and universities. Lots of homes have a LAN to connect various devices, such as PCs, tablets, smart TVs and printers.
Why use a LAN?
Sharing files is easier - network users can access the same files, work collobratively on them and copy files between machines.
You can share the same hardware, such as printers, on a LAN.
The Internet Connection can be shared between every device connected to the LAN.
You can install and update software on all computers at once, rather than one-by-one.
You can communicate with LAN users cheaply and easily, e.g. with instant messaging.
User accounts can be stored centrally, so users can log in from any device on the network.
Stands for Wide Area Network and connects LANs that are in different locations. For example, a business with offices in three different countries would need a WAN for all their devices to connect together.
Unlike a LAN, organisations hire infrastructure such as communication lines from telecommunications companies, who own and manage the WAN. This is because a WAN is much more expensive to set up than a LAN.
WANs may be connected using fibre or copper telephone lines, satellite links or radio links. The Internet is the biggest WAN.
Factors affecting the performance of networks
Bandwidth - the amount of data that can be transferred in a given time, e.g. 500 mbps. The greater the bandwidth, the better the network can perform.
Available bandwidth is shared between users of a network - too many users or heavy use such as streaming videos may cause congestion and slow the network. You can limit the bandwidth available to individual users to address this.
Wired connections are generally faster and more reliable than wireless. Fibre optic cables can give much better performance than copper cables. Wireless performance depends on signal quality so is affected by the range of the device, the amount of interference from other devices and physical obstructions.
Choice of hardware other than cables and network topology also have a big effect.
In a Star Topology, all the devices are connected to a central switch or server that controls the network. The central switch allows many devices to access the server simultaneously. Star networks may be wired or wireless.
Pros of a Star Topology: If a device fails or is disconnected, the rest of the network is unaffected. It's simple to add more devices to the network. Better performance than other setups - data goes straight to the central device so all devices can transmit data at once (unlike ring network) and there are very few data collisions (unlike bus network).
Cons of a Star Topology: In wired networks, every device needs a cable to connect to the central switch or server. This can be expensive, e.g. for an office building with 50 terminals. If there is a problem with the switch or server then the whole network is affected.
Bus Topology - all devices are arranged in a line, connected to a single backbone cable. Devices send data in both directions. This causes data collisions which slows the network.
Ring Topology - data moves in one direction around the ring, preventing collisions. But only one device can send data at a time and data passes through many devices before reaching its destination.
A mesh topology is a relatively new network layout. It is decentralised - networking devices are either directly or indirectly connected to every other one without the need for one central switch or server.
Mesh networks work by sending data along the fastest route from one device to another.
The main advantage of a mesh topology is that there is no single point where the network can fail. If the central switch or server of a star network fails then the whole network fails - in a mesh network, if one device fails then the data is sent along a different route to get to its target.
The traditional problem with mesh networks has been that they were very expensive - you needed a lot of wire to connect so many devices together. But now more people are using wireless technology, mesh networks are a more practical option.
A full mesh topology is where every device is connected to every other device. In a partial mesh topology, not all devices are fully-connected.
A protocol is a set of rules for how devices communicate and how data is transmitted across a network.
Protocols cover how communication between two devices should start and end, how the data should be organised, and what the devices should do if data goes missing.
MAC addresses are assigned to all network-enabled devices by the manufacturer. They are unique to the device and cannot be changed. MAC addresses are mainly used by the Ethernet protocol on LANs. LAN switches read the MAC addresses and use them to direct data to the right device.
IP addresses are used when sending data between TCP/IP networks e.g. over the internet. IP addresses are assigned either manually (static) or automatically (dynamic).
Static IP addresses - permanent addresses. They're used to connect printers on a LAN, and for hosting websites on the Internet - companies don't want the IP address of their website changing. Static IP addresses can be very expensive.
Dynamic IP addresses - assigned to the device by a network server, meaning your device may have a different IP address every time you log on to the network. ISPs commonly use dynamic IP addresses as they are more cost effective and can be reused.
TCP/IP is the protocol which dictates how data is sent between networks. It is made up of two protocols:
Transmission Control Protocol (TCP) - sets the rules for how devices connect on the network. It's in charge of splitting the data into packets and reassembling the packets back into the original data once it reaches the receiving device. It's also responsible for checking the data is correctly sent and delivered.
Internet Protocol (IP) is responsible for packet switching.
Other protocols you need to know...
HTTP stands for Hyper Text Transfer Protocol. Used by web browsers to access websites and communicate with web servers.
POP3 stands for Post Office Protocol V3. Used to retrieve emails from a server which are held by the server until downloaded at which point it is deleted.
FTP stands for File Transfer Protocol. Used to access, edit and move files between devices on a network.
IMAP stands for Instant Message Access Protocol. Used to retrieve emails from a server which are held by the server until you delete it. Used by most web-based email clients.
HTTPS stands for HTTP Secure. A more secure version of HTTP; encrypts all information sent and received.
SMTP stands for Simple Mail Transfer Protocol. Used to send emails and is also used to transfer emails between servers.
Layer 1 - Data Link Layer. Passing data over the physical network. Responsible for how bits are sent as electrical signals over cables, wireless and other hardware. Example is Ethernet.
Layer 2 - Network Layer. Making connections between networks, directing data packets and handling traffic. Used by routers. Example is IP.
Layer 3 - Transport Layer. Controlling data flow - e.g. splitting data into packets and checking packets are correctly sent and delivered. Example is TCP.
Layer 4 - Application Layer. Turning data into websites and other applications and vice versa. Examples include HTTP, FTP and SMTP.
Advantages of using layers
It breaks network communication into manageable pieces. This helps developers concentrate on only one area of the network without having to worry about the others.
As layers are self-contained, they can be changed without the other layers being affected.
Having set rules for each layer forces companies to make compatible, universal hardware and software, so different brands will work with each other and always work basically in the same way.
1) The sending device splits the data into packets to be sent across the network. Each packet is given a packet number to show the order of the data.
2) Each router reads the packet header and decides which way to send the packet next, according to the IP rules.
3) The way data is sent changes depending on network traffic so the packets can take different routes. If a router receives too many packets at once it may prioritise some over others.
4) As the packets take different routes, they can arrive in the wrong order. The receiving device uses the packet numbers to reassemble them in the right order.
5) Sometimes packets go missing in transit so the receiving device checks periodically that all packets have been received. If it hasn't received them within a certain time, it sends a timeout message back to the sending device.
6) If all the data is received and the checksums match, a receipt confirmation is sent to the sending device.
Packet switching is efficient because there are so many routes that data can take - packets can reach their receiving device even if there's heavy traffic.
Client-server and Peer-to-Peer Networks
Client-server networks are made up of a Server and Client
Peer-to-Peer networks don't use servers
3.Clients send requests to the server, client server relationship.
4.The server stores all data and may request a password before sending the data asked for.
2.Files and software are usually stored centrally on the server.
5.Most uses of the internet work on a client-server relationship e.g. web servers
1.Mananged by a server, the devices connected to the server are clients.
Pros and Cons
Pros
Cons
Server dependence - if the server goes down all clients lose access.
The server may become overloaded if too many clients are accessing it at once.
Expensive and needs a IT specialist to maintain.
Easier to install and update software
Easier to manage network security
Easier to preform back-ups
Servers are very reliable and are always on
Easier to keep track of files as they are stored centrally
2.Can store files on individual devices and then share them.
3.you may use a P2P network at home to share files between devices, or connect devices to a printer.
1.In a P2P networks all devices are equal connecting to each other .
Pros and Cons
Pros
Cons
Easy to maintain - you don't need any expertise or expensive hardware
No Dependence on server - if one device fails the whole network isn't lost
Copying files between devieces creates duplicate files, its easy to lose track of whats stored.
Peer machines are less reliable and data may be lost if one fails.
No centralised management - devices need things installed individually. Backups are more complicated.
machines are prone to slow when other devices access them.
Wireless Access Point (WAP)
A Wireless Access Point is basically a switch that enables devices to connect wirelessly.
A WAP device is required in order to set up a network.
WAPs are not to be confused with hotspots - hotspots are locations where you can connect to a WAP.
NICs, Switches and Routers
Network Interface Controller (NIC)
Switches
Routers
A NIC is an internal piece of hardware that allows a device to connect to a network.
Nowadays, NICs are built into the motherboard and exist for both wired and wireless connections.
Switches connect devices on a LAN - they receive data (in units called frames) from one device and transmit it to the device on the network with the correct MAC address.
Routers are responsible for transmitting data between networks - they're always connected to at least two networks.
Transmission Media
Ethernet Cables
Connect devices in a LAN. The most common Ethernet cables are CAT5e and CAT6.
They are 'twisted pair' cables, containing four pairs of copper wires which are twisted to reduce internal interference.
Fibre Optic Cables
Transmit data as light.They are high performance, expensive cables.
They don't suffer interference and can transmit data over large distances without loss of signal quality.
Coaxial Cables
Made of a single copper wire surrounded by a plastic layer for insulation and a metallic mesh which provides shielding from outside interference.
Packets contain:
Packet header (this contains the control information, i.e. destination address, source address and packet number).
Payload is what's in the envelope.
Checksum number - form of validation that the payload data hasn't been corrupted during transit.
System Security
Forms of attack
Passive - when someone monitors data travelling on a network and intercepts any sensitive information. They use hardware and software such as sniffers. Passive attacks are hard to detect. Prevent by data encryption.
Active - when someone attacks a network with malware. Usually more easily detected and can be defended using a firewall.
Insider - someone within an organisation exploits their network access to steal information.
Brute force - cracking passwords through trial and error. Uses automated software to produce hundreds of likely password combinations. Can be prevented by settings such as account locking after a certain number of attempts and using strong passwords.
DoS (Denial of Service) - where a hacker tries to stop users from accessing a part of a network or website. Most involve flooding the network with useless traffic making it extremely slow or completely inaccessible.
Malware
Malware is malicious software installed on someone's device without their knowledge or consent.
Viruses attach to certain files which can spread by copying infected files and activated by opening them.
Worms self-replicate without user userso they can spread quickly.
Trojans are malware designed as legitimate software. They don't replicate themselves, the user just installs them not realising their hidden purpose.
Typical actions include:
Deleting or modifying files.
Scareware
Locking files - ransomware encrypts all computer files and the user receives a message asking them for a large sum of money to get the decryption key.
Spyware - secretly monitors user actions such as key presses.
Rootkits - alter permissions giving malware and hackers administrator-level access to devices.
Threats posed to networks
click to edit