Unit 19: Computer Networking

A1 Network types and models Assignment 1 · A.P1

Types of network and why we need them

By the end you can: explain what LAN, WLAN, WAN, SAN, intranet, extranet, internet and cloud mean, and why an organisation would choose each one.

Why do we connect computers at all?

A network is two or more devices joined together so they can share things. Organisations build networks so people can communicate (email, chat, calls), share data (one copy of a file everyone can open), share resources (one printer or internet connection for 30 people) and work together from different places. Different jobs need different kinds of network, which is why there are several types.

Intranet, extranet, internet and cloud

These words describe who is allowed in rather than how big the network is. An intranet is a private website and set of services that only staff can reach, such as the HR pages or a room-booking system. An extranet opens part of that private system to trusted outsiders, for example suppliers checking orders. The internet is the public network of networks that anyone can join. Cloud services are computers, storage and software that you rent from a provider and reach over the internet, instead of buying and running them yourself.

Wired and wireless integration

Most real networks mix both. Wired Ethernet is faster, more reliable and harder to eavesdrop on, so it suits desktop PCs, servers and printers. Wireless gives freedom to move and avoids drilling through walls. An access point plugged into a switch joins the two: wireless devices connect to the access point, and the access point passes their traffic onto the wired LAN, so everyone can reach the same servers and printers.

Watch

A1 Network types and models Assignment 1 · A.P1, A.M1

Topologies and Ethernet standards

By the end you can: describe physical and logical topologies, and explain why IEEE 802 standards matter.

Physical topology: how the cables are laid out

Logical topology: how the data actually flows

The logical topology is the path the data takes and the rules it follows, which can differ from the cabling. A diagram showing where equipment sits and where cables run is a physical diagram; one showing IP addresses, VLANs and which ports connect to what is a logical diagram. Your Assignment 2 design needs both.

Standards: the IEEE 802 family

Devices from different makers only work together because they follow the same standards. The IEEE (Institute of Electrical and Electronics Engineers) writes the 802 family: 802.3 is wired Ethernet, 802.11 is Wi-Fi and 802.15 covers personal networks such as Bluetooth. The Wi-Fi Alliance tests products so they can carry the Wi-Fi logo.

2.4 GHz travels further and passes through walls better; 5 GHz is faster but has a shorter range. Access points close together should use different channels so they do not interfere.

A1 Network types and models Assignment 1 · A.P1, A.D1

Network models: peer-to-peer, client/server and thin client

By the end you can: describe the three network models and weigh up their strengths and weaknesses for different clients.

Peer-to-peer (P2P)

Every computer is equal: each one can share its own files or printer and use other computers' resources. There is no central server, so it is cheap and quick to set up. But each user manages their own sharing and passwords, backups are scattered, and it becomes hard to manage beyond about 10 computers. It suits a home or a very small office.

Client/server

One or more powerful servers provide services to the clients: central logins (a directory service), shared folders with permissions, printing, email and backups. It needs a server, a network operating system and someone to manage it, which costs more. In return it is secure, easy to back up and scales to thousands of users. If the server fails, services stop, so important servers are duplicated.

Thin client

A thin client is a cheap, simple device (sometimes just a screen, keyboard and small box) that sends key presses and mouse clicks to a powerful server, which runs the programs and sends back the picture. Nothing is stored on the device, so it is secure and cheap to run, and every desk gets the same, centrally updated setup. It depends completely on the network and the server: if either is slow or down, nobody can work, and it suits heavy graphics or offline work poorly.

Watch

A1 Network types and models Assignment 1 · A.P1

Modern trends: virtualisation, cloud, BYOD and SDN

By the end you can: explain the main modern networking trends and the opportunities and risks they bring.

Virtualisation

In the past each service (web, email, files) ran on its own physical server, which sat mostly idle; this waste is called server sprawl. Virtualisation uses a hypervisor to run several virtual machines (VMs) on one physical server, each with its own operating system. Fewer servers means less cost, power and space, and VMs can be copied, backed up and moved between hosts easily.

Cloud computing

Cloud providers run huge data centres and rent out computing, storage and software over the internet. A public cloud (AWS, Azure, Google Cloud) is shared with other customers; a private cloud is used by one organisation only; a hybrid cloud mixes the two; an edge cloud puts small servers close to users or IoT devices to cut delay.

Watch

A2 Network components Assignment 1 · A.P2, A.M1

Network hardware: devices, switches, routers, access points and cables

By the end you can: explain the job and key characteristics of each piece of network hardware.

End devices

An end device is where a message starts or ends: PCs, laptops, phones, tablets, printers, IP phones, smart TVs, servers and IoT sensors. Each has a network interface card (NIC), wired or wireless, with a unique MAC address burned in by the maker. When comparing end devices, look at how they connect (wired, wireless or both), the speed of their NIC, and whether they are mobile.

Switches

A switch connects devices inside one LAN. It works at layer 2 of the OSI model using MAC addresses: it learns which device is on each port and sends each frame only where it needs to go, instead of to everyone like an old hub. Characteristics to compare: number of ports (8, 24, 48), speed (100 Mbps, 1 Gbps, 10 Gbps), PoE (power over Ethernet for phones and APs), whether it is managed (VLANs, security, monitoring) and whether it can also route (a layer 3 switch).

Routers

A router connects different networks, for example your LAN to the internet or a head office to a branch. It works at layer 3 using IP addresses, keeps a routing table of known networks, and chooses the best path for each packet. Routes can be typed in by an administrator (static routing) or learned automatically from other routers with a routing protocol such as RIP or OSPF (dynamic routing). Home routers combine a router, a switch, an access point and a firewall in one box.

Wireless access points

An access point (AP) lets wireless devices join a wired network. It broadcasts a network name (SSID), uses an 802.11 standard, and should use strong security (WPA2 or WPA3). Compare APs on standard, speed, range, number of users, and whether they are managed centrally.

Connection media

A2 Network components Assignment 1 · A.P2

Network software: operating systems, tools and applications

By the end you can: explain the software that runs, monitors, troubleshoots and uses a network.

Networking systems software

A network operating system (such as Windows Server or Linux) provides services like logins, file sharing and DHCP. Network devices run their own system software too: Cisco switches and routers run IOS, configured through a command line. Other systems software includes firewall and security software, virtualisation and cloud management tools.

Network applications

These are programs that exist because of the network: email, web browsers, video calls and VoIP, shared databases (for example a stock or student records system), document management systems (shared editing with version history), remote access tools and cloud storage apps. When you plan a network, list the applications first, because they decide the bandwidth, servers and security you need.

A3 Communication standards and protocols Assignment 1 · A.M1

The OSI and TCP/IP models, ports and encapsulation

By the end you can: describe the OSI and TCP/IP layers, how data is encapsulated, and why protocols and port numbers matter.

A protocol is a set of rules that both ends agree on, such as how to address a message, how big each piece is and what to do if part goes missing. Layered models split the job of networking into smaller tasks, so each layer can be designed, understood and troubleshot on its own, and equipment from different makers can work together.

TCP or UDP?

Ports and sockets

An IP address gets data to the right device; a port number gets it to the right program on that device. A web server listens on port 443 for HTTPS, while the same server's email service listens on 25. An IP address plus a port, such as 192.168.1.10:443, is called a socket, which lets one computer hold many conversations at once.

Watch

A4 Infrastructure services and resources Assignment 1 · A.P2, A.M1

Network services: DNS, DHCP, directory, authentication and more

By the end you can: explain what each infrastructure service does and why a network needs it.

Hardware carries the data, but services make the network useful. Most run on servers, and in Assignment 2 you will set several of them up yourself.

DNS: names to numbers

DHCP: automatic addresses

Without DHCP an administrator would type an address into every device, and mistakes (two devices with the same address) would be common. DHCP hands out addresses from a pool (scope) for a set time (the lease). Servers, printers and network devices usually keep static addresses so they are always easy to find, while PCs, laptops and phones use DHCP.

B1 Design strategies and architectures Assignment 2 · B.P3, B.M2

Design strategies: goals, size, aims and constraints

By the end you can: explain why networks must be designed and planned, and use design aims to shape a design.

Start with the client, not the cables

A network exists to support an organisation's business goals (for example: open a second office, sell online, let staff work from home). A designer turns these into technical requirements: how many users and devices, which applications and services, how fast, how secure, and how much downtime is acceptable. Designing and planning first means the network is right-sized, secure and within budget, the build goes smoothly, and there is documentation for the people who will support it later.

Design aims

Every design has constraints (budget, time, the building itself, skills available) and involves trade-offs: more redundancy costs more; more security can make things harder to use. A good design explains which trade-offs were made and why.

Watch

B1 Logical network design Assignment 2 · B.P4

IP addressing, subnets, IPv6, naming and VLANs

By the end you can: plan an IP addressing scheme, choose between private/public and IPv4/IPv6, and use naming schemes and VLANs.

IPv4 addresses and subnet masks

An IPv4 address is 32 bits, written as four numbers from 0 to 255 (for example 192.168.10.77). Part of it identifies the network and the rest identifies the host (device). The subnet mask (for example 255.255.255.0, written /24) says where the split is. Every network has a network address (all host bits 0), a broadcast address (all host bits 1) and usable host addresses in between. The router's address on the LAN is the default gateway.

Private addresses are used inside organisations and are free to use, but they cannot travel across the internet. Public addresses are unique worldwide and are supplied by an ISP. NAT joins the two.

IPv6

IPv4 has only about 4.3 billion addresses, which have run out. IPv6 uses 128-bit addresses written in hexadecimal, giving about 3.4 × 10³⁸ addresses, so NAT is not needed. Two rules shorten them: leave out leading zeros in each group, and replace one run of all-zero groups with ::.

Naming schemes

A naming scheme gives every device a predictable name that says what and where it is, which makes documentation, troubleshooting and scripts much easier. For example LDN-FL2-SW01 (London, floor 2, switch 1), LDN-SRV-DC01 (London, server, domain controller 1) or BRN-PC-07. Avoid personal names, because people move.

VLANs

VLAN design issues to consider: VLANs add complexity, need managed switches, need a router or layer 3 switch for traffic between them, and a misconfigured port can put a device in the wrong VLAN, which is a security risk.

B2 Network development planning Assignment 2 · B.P4, C.P6

Choosing components and planning the installation

By the end you can: select hardware, software and services for a design, and plan device configuration and testing.

Selecting components

Choose each part by asking what does the client need it to do? Start from the services and users, then work outwards to servers, storage, switching, wireless and the WAN. Record the reason for every choice: this becomes your justification (B.M2).

Planning the configuration

Build a prototype in Packet Tracer first. Simulation is free, safe and lets you find mistakes before real equipment is bought or configured.

Planning the tests

A good test plan covers every requirement, says exactly how to test it, and states the expected result before you test. When you run it, add the actual result, a pass or fail, a screenshot reference, and what you did about any failure.

B3 Network services and resources access Assignment 2 · B.P4, C.P5

Users, groups, passwords and permissions

By the end you can: plan authentication, users and groups, and access permissions for files, folders and printers.

Authentication planning

A password policy sets the rules for passwords. Current UK NCSC guidance favours long passphrases (for example three random words), blocking common and breached passwords, multi-factor authentication (MFA) for important accounts, and not forcing frequent changes, which makes people choose weaker passwords. An audit policy decides what gets logged, such as successful and failed logins, changes to user accounts and access to sensitive folders, so problems can be investigated.

Users and groups

Give permissions to groups, not individuals, then put each user in the right groups. When someone joins, moves or leaves, you change their group membership instead of editing every folder. Use a consistent naming scheme for accounts (for example firstname.lastname or j.smith) and follow the principle of least privilege: people get only the access their job needs.

Windows uses NTFS permissions (Full control, Modify, Read and execute, Read, Write), which can be set per user or group. Linux uses the owner/group/others model above. Printers can also have permissions, for example only the MANAGERS group may use the colour printer.

sudo groupadd clinical
sudo useradd -m -G clinical tom.okafor
sudo passwd tom.okafor
sudo mkdir /srv/patients
sudo chown sam.berg:clinical /srv/patients
sudo chmod 770 /srv/patients
ls -ld /srv/patients

Creating a group, a user and a folder only that group can use (Linux)

Watch

C1 Implementation and configuration Assignment 2 · C.P5

Building and configuring the network

By the end you can: configure switches, routers, wireless and servers, and create users, groups and shared resources.

Build your prototype in Packet Tracer (or on real kit) in the same order every time: cable the devices, configure device security and names, set addresses, configure routing, add services, then add users and permissions. Save your configuration and take screenshots as you go: they are your evidence.

enable
configure terminal
hostname DEN-SW01
enable secret Str0ng-Enable!
line console 0
 password C0ns0le-Pass
 login
 exit
line vty 0 15
 password Vty-Pass-2026
 login
 exit
service password-encryption
banner motd # Authorised staff only #
end
copy running-config startup-config

Securing a switch: name, passwords on every way in, encrypted passwords and a warning banner

configure terminal
interface g0/0
 ip address 192.168.20.1 255.255.255.0
 no shutdown
 exit
ip dhcp excluded-address 192.168.20.1 192.168.20.20
ip dhcp pool SURGERY
 network 192.168.20.0 255.255.255.0
 default-router 192.168.20.1
 dns-server 192.168.20.2
 exit
ip route 192.168.30.0 255.255.255.0 10.0.0.2
end
show ip interface brief
show ip route

Router: interface address, a DHCP pool, and a static route to the branch network

vlan 10
 name STAFF
vlan 20
 name VOICE
interface range fa0/1-10
 switchport mode access
 switchport access vlan 10
interface range fa0/20-24
 shutdown
end
show vlan brief

Switch: create VLANs, put ports in them and shut unused ports

C2 Testing and troubleshooting Assignment 2 · C.P6, C.M3

Testing the network and fixing faults

By the end you can: carry out a test plan, record results with evidence, and troubleshoot faults methodically.

Three kinds of test

• Connectivity tests: can each device reach its gateway, other subnets, the branch and the internet? (ping, tracert)
• Service tests: do DHCP, DNS, web, email, VoIP and file sharing work? (ipconfig, nslookup, browser, email client)
• Access tests: can each user reach what they should, and are they blocked from what they should not? Are device passwords and Wi-Fi security working?

A troubleshooting method

Optimising

Testing often shows things to improve. Optimisation (C.M3) means changing the network to make it perform better, be more secure or be easier to use, then re-testing to prove the improvement: for example moving phones to a voice VLAN, shutting unused switch ports, adding SSH instead of Telnet, adding a second AP where the signal was weak, or reserving static addresses for printers.

C3 Performance monitoring Assignment 2 · C.M3, BC.D2

Baselines, monitoring and logs

By the end you can: establish a performance baseline, monitor the network, and use event logs.

A baseline is a measurement of how the network normally behaves when it is working well: typical bandwidth use, ping times, server CPU and memory, disk space. Take it soon after the network is built. Later measurements are compared with the baseline, so you can tell a real problem from a normal busy period and show whether an optimisation helped.

Review logs regularly, not just after something breaks: repeated failed logins can mean someone is guessing passwords, and repeated warnings can predict a failing disk.

C4–C5 Evaluation, review and professional skills Assignment 2 · C.P7, BC.D2, BC.D3

Evaluating the network and managing yourself

By the end you can: evaluate the design and the built network against the client's requirements, suggest future improvements, and evidence your own planning and professional behaviour.

Evaluate against the requirements

Go back to the client's requirements and judge each one with evidence from your design, build and tests. A review (Pass) says how far each requirement was met. An evaluation (Distinction) weighs up strengths and weaknesses of both the design and the implementation, uses evidence from every stage, explains why things happened, and reaches justified conclusions and future recommendations.

Future enhancements and optimisation

Suggest realistic next steps linked to the client's goals: capacity for more staff, redundancy (second internet line, a UPS), stronger security (MFA, network access control, meeting Cyber Essentials), better management (central monitoring, documentation), or new services (cloud backup, IPv6).

C5: skills, knowledge and behaviours