Why is "Confusing a switch with a router — saying a router 'connects devices in a LAN' or that a switch 'connects to the internet'." a common mistake in Networks including the internet?

Why it happens: Both forward data and a home 'router' box often contains a built-in switch, so students blur the two devices. How to avoid it: Lock in the split: a SWITCH works WITHIN one network and forwards frames using MAC addresses; a ROUTER connects DIFFERENT networks and forwards packets using IP addresses. Source: Examiner reports frequently note candidates mixing up switch and router roles.

Why is "Saying a switch 'sends data to all devices' — describing a switch as if it were a hub." a common mistake in Networks including the internet?

Why it happens: Students remember that a hub broadcasts and wrongly apply the same behaviour to a switch. How to avoid it: State the difference clearly: a hub broadcasts to ALL ports; a switch is selective and forwards a frame only to the destination device, found via its MAC address.

Why is "Describing CSMA/CD without the word 'random' — saying devices 'wait and try again' after a collision." a common mistake in Networks including the internet?

Why it happens: Students grasp the idea of retrying but omit that each device must wait a different, random time. How to avoid it: Always say the device waits a RANDOM (different) amount of time before retransmitting — otherwise the same two devices would just collide again. Source: A recurring loss of the final mark in CSMA/CD questions.

Why is "Treating the internet and the World Wide Web as the same thing." a common mistake in Networks including the internet?

Why it happens: In everyday speech people use 'the internet' and 'the web' interchangeably. How to avoid it: Remember: the internet is the global NETWORK/infrastructure of connected computers; the WWW is a SERVICE (web pages/resources) that runs on the internet.

Why is "Getting the IP address sizes/formats wrong — e.g. saying IPv4 is 64 bits, or that IPv6 is written in denary." a common mistake in Networks including the internet?

Why it happens: There are several numbers to remember (32/128 bits, octets vs hex groups) and they get muddled under exam pressure. How to avoid it: Memorise exactly: IPv4 = 32 bits = four denary octets (0–255); IPv6 = 128 bits = eight groups of four HEX digits separated by colons.

Why is "Answering 'justify which model/topology' questions with generic textbook facts that ignore the scenario." a common mistake in Networks including the internet?

Why it happens: Students revise the lists of benefits and drawbacks but don't practise linking them to the specific situation given. How to avoid it: For any 'justify for this situation' question, quote the requirement from the scenario (e.g. confidential central records) and explain why your chosen option meets it — context links earn the marks. Source: Examiner reports note that 'justify' marks are lost when answers are not applied to the given context.

Cambridge International A Levels Computer Science (9618)

Networks including the internet

0% Complete

Detailed Study Notes

Detailed notes on Communication for Cambridge International A Levels Computer Science, covering key concepts, explanations, examples, and exam-focused revision points.

Networks Including the Internet — Cambridge AS & A Level Computer Science 9618 Study Notes (2027-2029)

How computers are connected and why; LAN vs WAN; client-server vs peer-to-peer; thin vs thick clients; the bus, star, mesh and hybrid topologies; cloud computing; wired and wireless transmission media; the hardware that supports LANs and the internet; Ethernet and CSMA/CD; bit streaming; the internet vs the WWW; IP addressing and DNS. Most marks here come from precise definitions and well-justified choices for a given scenario.

At a glance

A network connects devices so they can share data, hardware and an internet connection and communicate.
LAN = small area (one site); WAN = large area (between cities/countries, external infrastructure).
Client-server: powerful central servers manage data/security; peer-to-peer: all computers are equals with no central server.
Thin-client depends on a central server for processing; thick-client processes locally and can work offline.
Topologies: bus (shared backbone), star (central switch), mesh (every node interlinked), hybrid (a mix).
Switch forwards frames by MAC address (within a LAN); router forwards packets by IP address (between networks).
Ethernet avoids clashes on a shared channel using CSMA/CD — listen, send, detect collision, wait a random time, retransmit.
The internet is the global network/infrastructure; the WWW is a service (web pages) that runs on it.
IPv4 = 32 bits (four denary octets 0–255); IPv6 = 128 bits (eight hex groups); DNS maps domain names → IP addresses.

What you’ll learn

Mapped to the Cambridge International A Level 9618 syllabus (2027-2029).

2.1 — Show understanding of the purpose and benefits of networking devices.
2.1 — Describe the characteristics of a LAN (local area network) and a WAN (wide area network).
2.1 — Describe and justify the use of the client-server and peer-to-peer models of networked computers, including thin-client and thick-client.
2.1 — Describe the bus, star, mesh and hybrid topologies, including how packets are transmitted between two hosts, and justify their use.
2.1 — Show understanding of cloud computing, including public and private clouds, and their benefits and drawbacks.
2.1 — Describe wired and wireless transmission media: copper cable, fibre-optic cable, radio waves (incl. WiFi), microwaves and satellites.
2.1 — Describe the hardware that supports a LAN (switch, server, NIC, WNIC, WAP, cables, bridge, repeater) and the role of a router.
2.1 — Show understanding of Ethernet and how collisions are detected and avoided using CSMA/CD.
2.1 — Show understanding of bit streaming (real-time and on-demand) and the importance of bit rate/broadband speed.
2.1 — Show understanding of the difference between the WWW and the internet, and the hardware/protocols that support the internet (modems, PSTN, dedicated lines, cell phone network).
2.1 — Show understanding of IP addressing (IPv4/IPv6, subnetting, public/private, static/dynamic), the URL and the role of the Domain Name Service (DNS).

Why network? LAN vs WAN

Networks let devices share resources; LANs are local, WANs are wide.

A network is two or more devices connected together so they can communicate and share resources. The benefits of networking are why almost every organisation does it:

Share hardware — one printer, scanner or server used by many devices.
Share data and files — work centrally rather than copying files by hand.
Share an internet connection across all devices.
Communicate — email, messaging, video calls.
Central management — software updates, backups and security can be applied in one place.

Networks are classified by the geographical area they cover:

Feature	LAN (Local Area Network)	WAN (Wide Area Network)
Area covered	Small — one site/building	Large — between cities/countries
Infrastructure	Usually owned by the organisation	Usually third-party/external
Example	School or office network	The internet; company branch links
Typical speed/cost	Faster, cheaper to run	Slower per link, higher cost

A WAN is often made by joining many LANs together over long-distance links.

Benefits: share hardware, data, internet; communicate; manage centrally.
LAN = small area, owner-controlled infrastructure.
WAN = large area, external/third-party infrastructure.
The internet is the largest WAN.

See the full worked example for networks including the internet →

Client-server vs peer-to-peer

Servers manage everything centrally; peers are all equal.

These are two ways of organising the computers on a network.

Client-server model. One or more powerful servers provide resources and services (files, printing, email, login/security). The other computers are clients that request these services.

Benefits: central storage, central backup, central security/access control, easier management, easier to add clients.
Drawbacks: servers are expensive; if a server fails, the service it provides is lost (single point of failure); needs IT expertise to run.

Peer-to-peer (P2P) model. Every computer is an equal peer: each can share its own files/resources and request resources from others, with no central server.

Benefits: cheap and simple to set up; no expensive server; no single point of failure.
Drawbacks: data is scattered across machines, so backup and security are hard to enforce; performance drops as more peers are added.

Justifying a choice is a common exam task: match the model to the scenario. Confidential data needing central backup and access control → client-server. A small home setup sharing files cheaply → peer-to-peer.

Client-server: central servers, central backup/security, but costly + single point of failure.
P2P: equal peers, cheap and simple, but scattered data and harder to secure.
Exam skill: justify the model FOR the given situation.

See the full worked example for networks including the internet →

Thin-client vs thick-client

Thin relies on the server; thick does its own processing.

Within a client-server setup, clients can be thin or thick depending on where the processing happens.

	Thin-client	Thick-client
Processing	Done mostly on the central server	Done locally on the device
Storage	Mostly on the server	Mostly local
Network	Needs a constant connection	Can work with little/no connection
Local resources	Limited (cheaper hardware)	Significant (more powerful hardware)
Management	Easy — update once on the server	Harder — each device managed separately

When a thin-client suits: many users doing similar, routine tasks (schools, call centres) where cheap hardware and central management/updates are valuable, and a reliable network is available.

When a thick-client suits: users needing to work offline or run heavy local applications (e.g. video editing on a laptop away from the network).

Thin-client: processing/storage on the server, needs constant connection, cheap to buy + easy to manage.
Thick-client: processing/storage local, works offline, more powerful hardware.
Difference is about WHERE the processing happens.

See the full worked example for networks including the internet →

Network topologies

Bus, star, mesh and hybrid — and how a packet travels in each.

A topology is the layout/arrangement of devices on a network. For each you should be able to describe how a packet gets from one host to another, and justify when it is suitable.

Bus. All devices share a single central cable (the backbone). When a host sends a packet, it travels along the backbone in both directions; every device sees it, but only the device with the matching destination address accepts it. Suits: small, cheap, temporary networks. Weakness: a backbone fault brings down the whole network, and only one device can transmit at a time (collisions).

Star. Every device has its own cable to a central device (a switch). A packet goes from the host to the central switch, which reads the destination address and forwards it only to the target device's cable. Suits: most modern LANs. Strength: one cable fault affects only one device; good performance with a switch. Weakness: if the central device fails, the whole network fails.

Mesh. Every node is connected to many (or all) other nodes, so packets can take multiple possible routes to the destination; if one link fails, another path is used. Suits: situations needing high reliability (e.g. the internet backbone). Weakness: lots of cabling/links → expensive (a full mesh of $n$ nodes needs many links).

Hybrid. A combination of two or more topologies (e.g. several star networks joined by a bus). Suits: large organisations that need flexibility, combining the strengths of each.

Bus: shared backbone; packet travels along it, only the addressed device accepts.
Star: central switch forwards each packet only to the target device.
Mesh: many interconnections → multiple routes → high reliability, costly.
Hybrid: a mix of topologies for flexibility.

See the full worked example for networks including the internet →

LAN hardware, routers and transmission media

Switch, NIC, WAP, bridge, repeater, router — and the cables/waves that carry data.

Hardware that supports a LAN:

Device	Role
Switch	Connects devices in a LAN; forwards a frame to a specific device using its MAC address
Server	A powerful computer that provides resources/services to clients
NIC	Network Interface Card — gives a wired connection and holds the device's MAC address
WNIC	Wireless NIC — connects a device to the network over radio waves
WAP	Wireless Access Point — lets wireless devices join a wired network (WiFi)
Cables	The physical wired connections (copper or fibre)
Bridge	Connects two LANs (or two segments) together so they act as one
Repeater	Boosts/regenerates a signal so it can travel a longer distance

Router (key device). A router connects different networks together (e.g. a home LAN to the internet) and forwards packets between networks using the destination IP address to choose the path. Compare: a switch works inside one network (MAC); a router works between networks (IP).

Transmission media (wired and wireless):

Medium	Type	Key points
Copper cable	Wired	Cheap, easy to install; signal weakens over distance; can suffer interference
Fibre-optic cable	Wired	Sends light; very fast, long distance, immune to electrical interference; more costly
Radio waves (WiFi)	Wireless	Convenient short-range mobility; lower speed/range, can be obstructed and intercepted
Microwaves	Wireless	High-frequency, line-of-sight links between dishes/towers; needs clear path
Satellites	Wireless	Global coverage over huge distances; high latency (long delay) and weather-affected

Switch (MAC, within LAN) vs router (IP, between networks) — the classic distinction.
NIC/WNIC give network connection + MAC; WAP provides WiFi access to a wired LAN.
Bridge joins two LANs; repeater boosts a signal over distance.
Fibre = fast/long/no interference but costly; copper = cheap but weaker; wireless = mobile but slower/interceptable.

See the full worked example for networks including the internet →

Ethernet and CSMA/CD

On a shared channel, devices listen before sending and recover from collisions.

Ethernet is the standard set of rules (protocols) used on wired LANs to control how data is sent. On a shared channel, two devices could transmit at once and their signals would collide, corrupting the data. Ethernet handles this with CSMA/CD — Carrier Sense Multiple Access with Collision Detection:

Carrier Sense — before sending, a device listens to the channel to check it is idle.
If the channel is busy, the device waits; if idle, it begins transmitting.
Collision Detection — the device keeps listening while it transmits. If two devices sent at once, a collision is detected.
On detecting a collision, the devices stop transmitting (and send a jam signal so all devices know).
Each device then waits a random (different) amount of time — a random backoff — before going back to step 1 and retransmitting.

The random wait is essential: it makes it unlikely the same two devices will collide again on the retry.

Ethernet = the wired-LAN rule set; CSMA/CD manages a shared channel.
Listen first (carrier sense); send if idle.
Detect collisions while transmitting; stop and jam.
Wait a RANDOM time, then retransmit.

See the full worked example for networks including the internet →

The internet, the WWW, and bit streaming

The internet is the network; the WWW is a service on it; streaming sends media as a continuous bit stream.

Internet vs World Wide Web — a very common 2-mark question:

The internet is the global network/infrastructure of interconnected computers and the hardware/cabling that links them.
The World Wide Web (WWW) is a service that runs on the internet: interlinked web pages and resources accessed via URLs and protocols such as HTTP/HTTPS in a browser.

Hardware that supports the internet:

Hardware	Role
Modem	Converts digital data to/from a form that can travel over telephone/cable lines (modulate/demodulate)
PSTN	Public Switched Telephone Network — the traditional phone network used to carry data
Dedicated (leased) lines	A permanent, private connection reserved for one organisation — reliable, consistent speed
Cell phone network	Provides mobile internet access over the mobile/cellular network

Bit streaming sends audio/video as a continuous stream of bits that plays as it arrives (rather than fully downloading first), using a buffer to store data slightly ahead of playback:

Real-time streaming — captured and sent live as it happens (live broadcast, video call); cannot pause/rewind; minimal buffering.
On-demand streaming — content is a pre-stored file on a server (e.g. a film); the user controls playback (pause, rewind, fast-forward); buffered ahead.

Why bit rate/broadband speed matters: the bit rate is the data transferred per second. If the broadband speed is lower than the stream's required bit rate, the buffer empties faster than it refills, so playback stalls ('buffering') or quality drops. A high, steady bit rate keeps the buffer supplied for smooth playback.

Internet = the network/infrastructure; WWW = web pages/resources (a service) on it.
Internet hardware: modem, PSTN, dedicated/leased lines, cell phone network.
Real-time = live, no pause; on-demand = pre-stored file, user controls playback.
Bit rate must meet the stream's needs or the buffer empties and playback stalls.

See the full worked example for networks including the internet →

IP addressing, URLs and DNS

Every device needs an IP address; DNS turns names into IP addresses.

An IP address uniquely identifies a device on a network so data can be routed to it.

IPv4 vs IPv6:

	IPv4	IPv6
Size	32 bits	128 bits
Written as	Four denary octets (0–255), dotted	Eight groups of four hexadecimal digits, colon-separated
Example	192.168.0.1	2001:0db8:85a3:0000:0000:8a2e:0370:7334
Why IPv6?	Limited number of addresses	Vastly more addresses (the world ran out of IPv4)

Subnetting splits a large network into smaller logical sub-networks (subnets), which improves organisation, performance and security.

How an IP is associated with a device: it can be assigned statically (manually fixed — stays the same, useful for servers) or dynamically (assigned automatically, e.g. by a DHCP server, and may change each time the device connects).

Public vs private IP addresses:

A public IP is unique across the whole internet and is routable on it.
A private IP is used only within a local network and is not routable on the public internet. Devices share one public IP via the router (NAT). Security implication: a private IP hides the internal device behind the router, so it is harder for an external attacker to address it directly.

URLs and DNS. A URL is the human-readable web address (protocol + domain + path), but computers locate servers by IP address. The Domain Name Service (DNS) is a system of servers that maps domain names to IP addresses:

The browser takes the domain name from the URL.
It asks a DNS server for that domain's IP address.
If unknown, the request is passed to other DNS servers until the mapping is found.
The IP address is returned, and the browser uses it to contact the correct web server and load the page.

IPv4 = 32 bits / four denary octets (0–255); IPv6 = 128 bits / eight hex groups.
Static IP = fixed (servers); dynamic IP = auto-assigned (DHCP) and may change.
Private IP = not routable externally → hides the device (security); public IP = routable on the internet.
DNS translates a domain name (from a URL) into the IP address the browser needs.

See the full worked example for networks including the internet →

Quick recap

Networks let devices share hardware, data and an internet connection; LAN = small area, WAN = large area.
Client-server = central servers (backup/security) but costly; P2P = equal peers, cheap but scattered/insecure.
Thin-client relies on the server for processing; thick-client processes locally and can work offline.
Topologies: bus (shared backbone), star (central switch forwards to target), mesh (multiple routes, reliable), hybrid (a mix).
Switch forwards by MAC within a LAN; router forwards by IP between networks; CSMA/CD: listen, send, detect, wait a random time, retry.
Internet = the network/infrastructure; WWW = web pages (a service) on it; streaming needs enough bit rate or the buffer empties.
IPv4 = 32 bits (four octets), IPv6 = 128 bits (eight hex groups); DNS maps domain names → IP addresses.

Memorise this

Verbatim phrases and definitions Cambridge mark schemes credit.

LAN = small geographical area; WAN = large geographical area.
Switch = forwards frames by MAC (within a LAN); Router = forwards packets by IP (between networks).
CSMA/CD = listen, transmit if idle, detect collision, stop, wait a RANDOM time, retransmit.
Internet = network/infrastructure; WWW = service (web pages) running on the internet.
IPv4 = 32 bits = four denary octets 0–255; IPv6 = 128 bits = eight hex groups.
Private IP = not routable on the internet (hides device); public IP = routable.
DNS = translates domain names into IP addresses.

How it’s examined

Section 2.1 is assessed in Paper 1 (Theory Fundamentals, 1 h 30 min, 75 marks) — written, externally assessed, no calculator. Expect: short 'state/describe' questions (LAN vs WAN, switch vs router, internet vs WWW); 'describe how a packet is transmitted' for a named topology; a CSMA/CD sequence; and 'justify/recommend for this situation' questions on client-server vs P2P, topology choice, cloud, or transmission medium. The marks reward precise terminology (MAC vs IP, carry/random backoff, 32/128 bits) and context-linked justification.

Sources: Cambridge International AS & A Level Computer Science 9618 syllabus (2027-2029), section 2.1. Last reviewed 2026-06-21.

Take this whole topic with you

Step-by-step worked examples — Networks including the internet

Step-by-step solutions to past-paper-style questions on networks including the internet, written exactly the way a tutor would explain them at the board.

Question type:

1Identify LAN vs WAN features
Getting startedIdentify & classify• 9618 Paper 1 style• LAN, WAN, AO1
Question
A school has computers connected together within one building. The school also connects to its other branch in another city. State whether each connection is a LAN or a WAN, and give one reason for each. (4 marks)
Step-by-step solution
1. Step 1
  Decide the key discriminator: a LAN covers a small geographical area (one site/building) and usually uses cabling/equipment the owner controls; a WAN covers a large geographical area (between cities/countries) and usually uses third-party/external infrastructure.
2. Step 2
  Computers within one building cover a small area on equipment the school owns → this is a LAN.
3. Step 3
  The link to a branch in another city covers a large geographical area and relies on external telecommunication infrastructure → this is a WAN.
Answer
Within the building = LAN (small geographical area / school-owned infrastructure). To the branch in another city = WAN (large geographical area / uses external/third-party infrastructure).
Examiner tip
1 mark for each correct classification + 1 mark for each valid reason. The discriminator examiners want is geographical area (small vs large), with ownership of infrastructure as a strong supporting point.
2How a switch differs from a hub
Getting startedIdentify & classify• 9618 Paper 1 style• switch, hub, MAC, AO1
Question
Explain how a switch handles an incoming data frame differently from a hub. (3 marks)
Step-by-step solution
1. Step 1
  Recall what a hub does: it has no intelligence, so it simply broadcasts (copies) every incoming frame out of all of its ports.
2. Step 2
  Recall what a switch does: it reads the destination MAC address in the frame and looks it up in its MAC address table.
3. Step 3
  Using the table, the switch forwards the frame out of only the single port connected to the destination device, rather than to every device.
Answer
A hub broadcasts every frame out of all its ports to all devices. A switch reads the frame's destination MAC address, finds the matching port in its MAC address table, and forwards the frame to only that one device.
Examiner tip
The high-value point is MAC address: a switch is selective because it uses the destination MAC address. Answers that only say 'a switch is faster' without explaining the MAC-based forwarding cap the marks.
3Justify a star topology over a bus topology
Building confidenceWord problem• 9618 Paper 1 style• topology, star, bus, justify, AO2
Question
An office is setting up a new wired network for 12 computers and is choosing between a bus topology and a star topology. Recommend which topology the office should use, justifying your choice with two benefits and one drawback. (6 marks)
Step-by-step solution
1. Step 1
  Recall the layouts: in a bus topology every device connects to a single shared backbone cable; in a star topology every device has its own dedicated cable to a central device (a switch).
2. Step 2
  Benefit (reliability): in a star, a single cable fault affects only the one device on that cable, so the rest of the network keeps working. In a bus, a fault/break in the backbone cable can bring down the whole network — so star is more reliable for the office.
3. Step 3
  Benefit (performance): a star uses a switch that sends each frame only to the destination device, so there are no shared-cable collisions; a bus shares one cable so collisions and contention slow it down as traffic grows. With 12 computers the star performs better.
4. Step 4
  Drawback (cost): a star needs a separate cable for every device plus a central switch, so it uses more cabling and hardware than a bus, which just taps onto one backbone cable — making star more expensive to install.
5. Step 5
  Conclude with a justified recommendation: choose star because for 12 office computers reliability and performance outweigh the higher cabling cost.
Answer
Recommend a STAR topology. Benefit 1: each computer has its own dedicated cable to a central switch, so if one cable fails only that computer is affected and the rest of the network keeps working (a bus fails entirely if the backbone breaks). Benefit 2: the switch sends each frame only to its destination, avoiding the shared-cable collisions of a bus, so performance is better as traffic grows. Drawback: a star needs more cabling and a central switch, so it costs more to install than a bus that shares one backbone cable. Overall, for 12 office computers the reliability and performance of a star justify the extra cost.
Examiner tip
This is the 'justify topology choice' type that appears every year. Marks for: two developed benefits of the chosen topology vs the alternative (up to 2 each) + one valid drawback (up to 2). You must compare against the OTHER topology and reach a recommendation — listing isolated facts without a justified choice loses marks.
4Explain CSMA/CD collision handling
Building confidenceMulti-step problem• 9618 Paper 1 style• Ethernet, CSMA/CD, collision, AO1
Question
An Ethernet network uses CSMA/CD. Describe the steps CSMA/CD takes when two devices try to transmit at the same time. (5 marks)
Step-by-step solution
1. Step 1
  Carrier Sense: before transmitting, a device listens to the shared channel to check whether it is idle.
2. Step 2
  If the channel is busy the device waits; if it is idle the device begins to transmit its data.
3. Step 3
  Collision Detection: the device keeps listening while transmitting. If two devices transmit at once, their signals collide; the devices detect this collision.
4. Step 4
  On detecting a collision the devices stop transmitting (and may send a jam signal so all devices know a collision occurred).
5. Step 5
  Each device then waits a random (different) amount of time — a random backoff — before listening again and re-transmitting, which reduces the chance of the same two colliding again.
Answer
Listen to the channel (carrier sense). 2) If idle, begin transmitting; if busy, wait. 3) Keep listening while sending; if a collision is detected the data is corrupted. 4) Stop transmitting (send a jam signal). 5) Each device waits a random/different amount of time (random backoff) before checking the channel and retransmitting.
Examiner tip
Award marks across the sequence: listen before sending (1), transmit if idle (1), detect collision while transmitting (1), stop/abandon transmission (1), wait a RANDOM time before retrying (1). The word 'random' is essential — if both waited the same fixed time they would simply collide again.
5Justify client-server vs peer-to-peer
StretchWord problem• 9618 Paper 1 style• client-server, peer-to-peer, justify, AO2
Question
A medium-sized company stores confidential customer records that must be centrally backed up and have controlled access. Recommend whether a client-server or peer-to-peer model is more suitable, justifying your choice with two reasons. (4 marks)
Step-by-step solution
1. Step 1
  Identify the requirements: central storage, central backup, and controlled (managed) access to confidential data.
2. Step 2
  Recall the models: in client-server, dedicated servers hold data and manage logins/security centrally; in peer-to-peer, each computer is an equal that shares its own resources with no central control.
3. Step 3
  Match to requirements: client-server allows data and security to be managed in one place — centralised backup of records and centralised access control (user accounts/permissions) — which exactly meets the company's needs.
4. Step 4
  Justify why P2P is unsuitable: data would be scattered across many machines, making consistent backup and controlled access hard to enforce, increasing the security risk for confidential records.
Answer
Recommend client-server. Reason 1: data is stored centrally on a server so it can be backed up reliably and consistently in one place. Reason 2: access/security is managed centrally (user accounts and permissions on the server), so access to confidential records can be controlled — peer-to-peer would scatter data across machines, making backup and access control difficult and less secure.
Examiner tip
This is a 'justify for a situation' question — you must link features to THIS scenario (confidential records, central backup, controlled access). Two developed, context-linked reasons earn full marks; generic 'servers are better' answers do not.
6Compare IPv4 and IPv6, and public vs private IP
StretchWord problem• 9618 Paper 1 style• IP address, IPv4, IPv6, AO1, AO2
Question
(a) State the size in bits of an IPv4 address and of an IPv6 address, and describe the format of each. (b) Explain one security advantage of a device on a home network using a private IP address. (5 marks)
Step-by-step solution
1. Step 1
  IPv4 is 32 bits, written as four denary numbers (octets) separated by dots, each in the range 0–255 (dotted-decimal).
  $\text{IPv4} = 32\text{ bits} = 4 \times 8\text{-bit octets}$
2. Step 2
  IPv6 is 128 bits, written as eight groups of four hexadecimal digits separated by colons.
  $\text{IPv6} = 128\text{ bits} = 8 \times 16\text{-bit groups (hex)}$
3. Step 3
  A private IP address is only used inside the local network and is not routable on the public internet, so the device cannot be addressed directly from outside.
4. Step 4
  Because external computers cannot reach the device directly (the router/NAT hides it behind one public IP), it is harder for an attacker on the internet to target that specific device — a security advantage.
Answer
(a) IPv4 = 32 bits, written as four denary octets (0–255) separated by dots, e.g. 192.168.0.1. IPv6 = 128 bits, written as eight groups of four hexadecimal digits separated by colons. (b) A private IP address is not routable on the public internet, so the device is hidden behind the router's single public IP and cannot be addressed directly from outside, making it harder for an external attacker to target.
Examiner tip
(a) Exact figures matter: 32 bits / four octets / 0–255 for IPv4, and 128 bits / eight hex groups for IPv6. (b) The security point is that a private IP is not routable externally / the device is hidden behind the public IP — link this to it being harder to attack, not just 'it is private'.

Model Answers — Networks including the internet

High-scoring sample answers for networks including the internet on the Cambridge IGCSE paper, with examiner-style notes mapping each response to the mark scheme and assessment objectives.

Question 1
9618 Paper 1 style2 marks
State the difference between the internet and the World Wide Web (WWW). (2 marks)
Model answer
The internet is the global network of interconnected computers/networks and the hardware and infrastructure that connect them.

The World Wide Web (WWW) is a service that runs on the internet: a collection of interlinked web pages and other resources, accessed using URLs and protocols such as HTTP/HTTPS through a web browser.
Why this scores
1 mark for the internet = the network/infrastructure of connected computers; 1 mark for the WWW = the web pages/resources (a service) that use the internet. The key idea is that the WWW is one service that runs on the internet, not the internet itself.
Question 2
9618 Paper 1 style4 marks
(a) Distinguish between a static IP address and a dynamic IP address. (b) State the purpose of subnetting a network. (4 marks)
Model answer
(a) Static IP address: an IP address that is fixed/permanently assigned to a device and does not change between connections. It is set up manually (or reserved) and is useful for devices that must always be found at the same address, such as a web server.

Dynamic IP address: an IP address that is assigned automatically to a device when it joins the network (typically by a DHCP server) and may be different each time it connects. This reuses a pool of addresses so fewer addresses are needed than there are devices.

(b) Purpose of subnetting: subnetting divides a single large network into smaller logical sub-networks (subnets). This reduces network traffic/congestion (broadcasts are kept within a subnet), improves performance and security (subnets can be isolated/managed separately), and makes the network easier to manage.
Why this scores
(a) 1 mark for static = fixed/manually assigned address that stays the same; 1 mark for dynamic = automatically assigned (DHCP) and can change each connection. (b) Up to 2 marks for purpose of subnetting: splits one network into smaller subnets to reduce traffic/congestion and improve performance, security or manageability. 'Static vs dynamic' and 'subnetting' are part of the IP-addressing topic examined every year.
Question 3
9618 Paper 1 style4 marks
Explain the difference between a thin-client and a thick-client, giving one situation where a thin-client would be appropriate. (4 marks)
Model answer
Thin-client: a device that relies heavily on a powerful central server to do most of the processing and storage; it has limited local resources and needs a constant network connection to the server to function.

Thick-client: a device that can do most of its own processing and storage locally; it has its own significant resources and can work even with little or no network connection.

Appropriate situation: a school or call centre with many workstations doing routine tasks — thin-clients are cheaper to buy and easier to maintain/update centrally (all processing and software are managed on the server), so they suit large numbers of users doing similar work.
Why this scores
Marks for: thin-client depends on the server for processing/storage (1), thin-client needs a constant connection / limited local power (1), thick-client does its own processing/storage and can work offline (1), a valid, justified situation (1). The contrast must be about WHERE processing happens.
Question 4
9618 Paper 1 style6 marks
A business is considering storing its data using public cloud computing. Discuss two benefits and one drawback of this choice. (6 marks)
Model answer
Benefit 1 — access and scalability: the data is stored on the cloud provider's servers and can be accessed from anywhere with an internet connection, on any authorised device. Storage can be scaled up or down on demand, and the business pays only for what it uses, avoiding the cost of buying and maintaining its own large servers.

Benefit 2 — backup and maintenance: the cloud provider takes responsibility for backups, redundancy and hardware maintenance, so data is less likely to be lost through a single hardware failure and the business needs less in-house IT expertise.

Drawback — dependence and security/privacy: access depends on a reliable internet connection, so if the connection (or the provider) goes down the business cannot reach its data. With a public cloud the data is held on shared, third-party infrastructure, raising security and privacy concerns and meaning the business must trust the provider to protect confidential data — a private cloud would give more control but at higher cost.
Why this scores
Each benefit and the drawback should be a developed point (statement + explanation), worth up to 2 marks each. Strong answers name the public/private distinction and tie the drawback to internet dependence or to data being on shared third-party servers (security/privacy).
Question 5
9618 Paper 1 style6 marks
Explain the difference between real-time and on-demand bit streaming, and explain why a sufficient bit rate (broadband speed) is important for streaming video. (6 marks)
Model answer
Real-time streaming: the data is captured and transmitted live as the event happens (for example a live sports broadcast or video call). It cannot be paused or rewound because the content is being produced in real time, and there is very little buffering.

On-demand streaming: the content already exists as a file stored on a server (for example a film on a streaming service). The user can choose when to start it and can pause, rewind and fast-forward; the data is buffered ahead before playback.

Why bit rate matters: the bit rate is the amount of data transferred per second. Video needs a high, steady bit rate to deliver enough frames smoothly. If the broadband speed is lower than the video's bit rate, the buffer empties faster than it fills, so playback stalls/pauses ('buffering'), the picture quality is reduced, or frames are dropped. A sufficient bit rate keeps the buffer supplied so playback is smooth and uninterrupted.
Why this scores
Marks for: real-time = live as it happens / limited buffering (up to 2); on-demand = pre-stored file, user controls playback / buffered ahead (up to 2); bit rate explanation = data per second must meet/exceed the stream's needs or the buffer empties and playback stalls/quality drops (up to 2). The buffer argument is the discriminating point.
Question 6
9618 Paper 1 style6 marks
A user types a URL into a web browser. Explain the role of the Domain Name Service (DNS) in allowing the browser to load the correct web page. (6 marks)
Model answer
A URL (Uniform Resource Locator) is a human-readable web address, e.g. https://www.example.com/page. Computers, however, locate servers using IP addresses, not domain names.

The Domain Name Service (DNS) is a system of servers that holds a directory mapping domain names to IP addresses. When the user enters a URL:

The browser extracts the domain name from the URL.

A request is sent to a DNS server asking for the IP address of that domain.

If that DNS server does not hold the mapping, the request is passed on to other DNS servers until the matching IP address is found.

The IP address is returned to the browser.

The browser then uses that IP address to contact the correct web server and request the web page.

In short, DNS translates the easy-to-remember domain name into the IP address the computer needs to locate the server.
Why this scores
Marks for: URL contains a domain name (1); computers need an IP address to locate a server (1); DNS holds/maps domain names to IP addresses (1); browser queries a DNS server (1); request may be forwarded to other DNS servers if not found (1); IP returned so the browser can contact the correct server (1). The core creditable idea is 'translates domain name → IP address'.

Key Definitions and Keywords — Networks including the internet

Definitions to memorise and the exact keywords mark schemes credit for networks including the internet answers — sharpened from recent examiner reports for the 2026 Cambridge IGCSE sitting.

LAN (Local Area Network)
Examiner keyword
A network that connects devices over a small geographical area, such as a single building or site, usually using infrastructure owned and managed by the organisation.
Example
Computers, printers and a server linked together within one school building.
WAN (Wide Area Network)
Examiner keyword
A network that connects devices over a large geographical area, such as between cities or countries, usually using external/third-party telecommunication infrastructure.
Example
The internet; or links between a company's offices in different cities.
Client-server model
Examiner keyword
A network model in which one or more powerful central computers (servers) provide resources and services, and other computers (clients) request them. Security, storage and processing are managed centrally.
Peer-to-peer (P2P) model
Examiner keyword
A network model in which every computer is an equal (a peer) that can both share and request resources directly, with no dedicated central server controlling the network.
Thin-client
Examiner keyword
A device that depends on a central server for most of its processing and storage and needs a constant network connection; it has limited local resources of its own.
Thick-client
A device that carries out most of its own processing and storage locally and can function with little or no network connection.
Switch
Examiner keyword
A device that connects devices within a LAN and forwards a data frame to a specific device by reading its destination MAC address (unlike a hub, which broadcasts to all devices).
Router
Examiner keyword
A device that connects different networks together and forwards data packets between them, choosing the path using the destination IP address.
NIC / WNIC
A Network Interface Card gives a device a wired connection to a network and holds its MAC address. A Wireless NIC (WNIC) does the same job over a wireless (radio) connection.
Wireless Access Point (WAP)
A device that allows wireless-enabled devices to connect to a wired network using radio waves (WiFi), acting as a bridge between wireless devices and the LAN.
CSMA/CD
Examiner keyword
Carrier Sense Multiple Access with Collision Detection — the Ethernet method where a device listens before sending, detects collisions while transmitting, then waits a random time before retransmitting.
IP address
Examiner keyword
A unique numerical address assigned to a device on a network so that data can be sent to and from it. IPv4 uses 32 bits (four denary octets); IPv6 uses 128 bits (eight hexadecimal groups).
Example
IPv4: 192.168.0.1 — four octets, each 0–255.
Domain Name Service (DNS)
Examiner keyword
A system of servers that translates human-readable domain names (from URLs) into the IP addresses computers use to locate the correct server.
URL (Uniform Resource Locator)
A human-readable address that identifies the location of a resource on the web, including the protocol, domain name and path, e.g. https://www.example.com/page.

Common Mistakes and Misconceptions — Networks including the internet

The traps other students keep falling into on networks including the internet questions — taken from recent Cambridge IGCSE examiner reports and mark schemes — and how to avoid them.

✕Confusing a switch with a router — saying a router 'connects devices in a LAN' or that a switch 'connects to the internet'.
Examiner reports frequently note candidates mixing up switch and router roles.
Why it happens
Both forward data and a home 'router' box often contains a built-in switch, so students blur the two devices.
How to avoid it
Lock in the split: a SWITCH works WITHIN one network and forwards frames using MAC addresses; a ROUTER connects DIFFERENT networks and forwards packets using IP addresses.
✕Saying a switch 'sends data to all devices' — describing a switch as if it were a hub.
Why it happens
Students remember that a hub broadcasts and wrongly apply the same behaviour to a switch.
How to avoid it
State the difference clearly: a hub broadcasts to ALL ports; a switch is selective and forwards a frame only to the destination device, found via its MAC address.
✕Describing CSMA/CD without the word 'random' — saying devices 'wait and try again' after a collision.
A recurring loss of the final mark in CSMA/CD questions.
Why it happens
Students grasp the idea of retrying but omit that each device must wait a different, random time.
How to avoid it
Always say the device waits a RANDOM (different) amount of time before retransmitting — otherwise the same two devices would just collide again.
✕Treating the internet and the World Wide Web as the same thing.
Why it happens
In everyday speech people use 'the internet' and 'the web' interchangeably.
How to avoid it
Remember: the internet is the global NETWORK/infrastructure of connected computers; the WWW is a SERVICE (web pages/resources) that runs on the internet.
✕Getting the IP address sizes/formats wrong — e.g. saying IPv4 is 64 bits, or that IPv6 is written in denary.
Why it happens
There are several numbers to remember (32/128 bits, octets vs hex groups) and they get muddled under exam pressure.
How to avoid it
Memorise exactly: IPv4 = 32 bits = four denary octets (0–255); IPv6 = 128 bits = eight groups of four HEX digits separated by colons.
✕Answering 'justify which model/topology' questions with generic textbook facts that ignore the scenario.
Examiner reports note that 'justify' marks are lost when answers are not applied to the given context.
Why it happens
Students revise the lists of benefits and drawbacks but don't practise linking them to the specific situation given.
How to avoid it
For any 'justify for this situation' question, quote the requirement from the scenario (e.g. confidential central records) and explain why your chosen option meets it — context links earn the marks.

Networks including the internet

Detailed Study Notes

At a glance

What you’ll learn

Quick recap

Memorise this

How it’s examined

Take this whole topic with you

1Identify LAN vs WAN features

2How a switch differs from a hub

3Justify a star topology over a bus topology

4Explain CSMA/CD collision handling

5Justify client-server vs peer-to-peer

6Compare IPv4 and IPv6, and public vs private IP

LAN (Local Area Network)

WAN (Wide Area Network)

Client-server model

Peer-to-peer (P2P) model

Thin-client

Thick-client

Switch

Router

NIC / WNIC

Wireless Access Point (WAP)

CSMA/CD

IP address

Domain Name Service (DNS)

URL (Uniform Resource Locator)

✕Confusing a switch with a router — saying a router 'connects devices in a LAN' or that a switch 'connects to the internet'.

✕Saying a switch 'sends data to all devices' — describing a switch as if it were a hub.

✕Describing CSMA/CD without the word 'random' — saying devices 'wait and try again' after a collision.

✕Treating the internet and the World Wide Web as the same thing.

✕Getting the IP address sizes/formats wrong — e.g. saying IPv4 is 64 bits, or that IPv6 is written in denary.

✕Answering 'justify which model/topology' questions with generic textbook facts that ignore the scenario.