Spec checklist
Rate each point. Be honest: 'Secure' means you could answer an exam question on it now without notes.
| 7.1.1 | Explain the features and uses of personal computers, mobile devices, servers and embedded devices |
| 7.1.2 | Explain input/output devices, processors (cores, clock speed, cache, mobile), RAM and ROM, secondary storage, motherboard, GPU, network interfaces, cooling and sensors |
| 7.2.1 | Explain batch, multitasking, real-time, network and mobile operating systems |
| 7.2.2 | Explain file management, defragmenters, compression, package managers, protection and backup utilities |
| 7.2.3 | Explain IDEs (editing, debugging, screen design tools), compilers and interpreters |
| 7.2.4 | Explain word processors, spreadsheets, databases, email and project management software |
| 7.3.1–2 | Explain the benefits and drawbacks of networks and the features of PAN, LAN, MAN, WAN and VPN |
| 7.3.3–4 | Compare copper, fibre and wireless connections; star, mesh and tree topologies; logical vs physical |
| 7.3.5–6 | Compare client-server, thin client and peer-to-peer; explain the role of servers, clients, routers, switches and the internet backbone |
| 7.3.7–8 | Describe the layers, functions and protocols of the OSI and TCP/IP models |
| 7.3.9 | Explain packet structure, packet switching, causes of packet loss and CRC |
| 7.3.10 | Explain HTTP/HTTPS, SMTP/POP/IMAP, RIP/OSPF, FTP/SFTP, DHCP and DNS |
| 7.3.11 | Explain bandwidth and latency and their effect on performance |
| 7.4.1–2 | Explain virtual machines, type 1 and type 2 hypervisors, and the key features of virtual environments |
| 7.4.3–4 | Explain the benefits and drawbacks of virtual environments |
| 7.5.1–2 | Explain private and public cloud and the benefits of cloud |
| 7.5.3 | Explain IaaS, PaaS and SaaS, including who manages what |
| 7.6.1 | Explain the benefits of resilient environments |
| 7.6.2 | Explain methods to improve resilience: updates, hardware replacement, redundancy, hardening, backups, hot/warm/cold sites, standard operating procedures |
7.1 Hardware
Types of physical computer (7.1.1)
| Type | Features | Typical uses |
|---|---|---|
| Personal computer (desktop, laptop) | General purpose; keyboard, mouse, screen; upgradeable (desktops); moderate power use | Office work, software development, gaming, design |
| Mobile devices (smartphones, tablets) | Portable, battery powered, touchscreen, built-in sensors (GPS, camera, accelerometer), mobile processors, wireless connectivity | Communication, apps, mobile payments, field work (e.g. delivery drivers) |
| Servers | Powerful multi-core processors, large RAM and storage, redundant power supplies and drives, run 24/7, often rack-mounted, no screen needed | Hosting websites, databases, email, file storage, virtual machines |
| Embedded devices | Built into a larger product, one dedicated function, low power, often a microcontroller with firmware in ROM, little or no user interface | Washing machines, car engine management, smart thermostats, medical devices |
Hardware devices (7.1.2)
| Component | Key points |
|---|---|
| Input devices | Send data into the computer: keyboard, mouse, touchscreen, microphone, scanner, barcode reader, webcam. |
| Output devices | Present results: monitor, printer, speakers, projector, actuators (motors in robots). |
| Processor: number of cores | Each core can process instructions independently. More cores help multitasking and programs written to run in parallel. Doubling cores does not double speed if software is not designed for it. |
| Processor: clock speed | Cycles per second in GHz (3.5 GHz = 3.5 billion cycles per second). Higher speed means more instructions per second but more heat and power. |
| Processor: cache size | Very fast memory inside the CPU (levels L1, L2, L3). Larger cache means fewer slow trips to RAM. |
| Processor: mobile processors | Designed for low power and low heat to save battery (e.g. ARM-based). Often a system on a chip (SoC) combining CPU, GPU and other components. |
| RAM | Volatile. Holds the OS, programs and data in use. More RAM lets more programs run at once without slowing down (less use of virtual memory). |
| ROM | Non-volatile. Holds firmware such as the BIOS/UEFI used to start the computer. Contents are not normally changed. |
| Secondary storage: magnetic | Hard disk drives (HDD): spinning platters and a moving read/write head. Cheap per GB, large capacity; slower, moving parts can fail if dropped, noisy. |
| Secondary storage: solid state | SSDs, USB flash drives, SD cards: flash memory, no moving parts. Fast, robust, silent, low power; higher cost per GB, limited write cycles. |
| Secondary storage: optical | CD, DVD, Blu-ray: read by laser. Cheap, portable, good for distribution and archiving; low capacity, slow, easily scratched, becoming obsolete. |
| Motherboard | Main circuit board connecting the CPU, RAM, storage, expansion cards and ports, and carrying data between them. |
| Graphics processing unit (GPU) | Thousands of small cores for parallel processing. Renders graphics; also used for machine learning, video editing and scientific calculations. |
| Network interface devices | Connect the computer to a network. Can be an internal PCI/PCIe card (fast, permanent, desktops and servers) or a USB adapter (portable, easy to add, may be slower). |
| Cooling: air | Heat sinks and fans. Cheap and simple; noisier and less effective under heavy load. |
| Cooling: liquid | Liquid pumped through a block on the CPU/GPU to a radiator. Quieter and more effective for high-performance machines and data centres; more expensive, risk of leaks. |
| Sensors | Measure the physical world: temperature, light, motion, GPS, accelerometer, fingerprint. Used in phones, IoT devices and embedded systems. |
★ EXAM TIP
When recommending hardware, link the specification to the use: 'A video editing workstation needs a GPU with many cores because rendering is done in parallel, which reduces export times.'
- Server
- A computer that provides services or resources (files, web pages, email) to other computers on a network.
- Core
- An independent processing unit within a CPU. More cores means more instructions can be processed at the same time.
- Clock speed
- The number of instruction cycles a CPU can carry out per second, measured in GHz.
- Cache
- Small, very fast memory on or near the CPU that stores frequently used instructions and data.
- RAM
- Random access memory: volatile main memory that holds the programs and data currently in use.
- ROM
- Read-only memory: non-volatile memory that holds the startup instructions (firmware/BIOS).
- Volatile
- Loses its contents when the power is switched off.
- GPU
- Graphics processing unit: a processor with many cores designed for parallel processing of graphics and large calculations.
- Embedded device
- A computer built into a larger device to perform one dedicated function, e.g. in a washing machine or car.
7.2 Software
Operating systems (7.2.1)
| Type | Features | Examples of use |
|---|---|---|
| Batch | Runs jobs without user interaction; processes high volumes of similar data; jobs scheduled to run at set times (often overnight). | Payroll runs, bank statement printing, end-of-day billing |
| Multitasking | Concurrent execution of multiple tasks; time-slicing gives each task a turn on the CPU; interrupts let urgent tasks take priority. | Windows, macOS, Linux desktops |
| Real-time (RTOS) | Responds to inputs within a guaranteed time. Used for monitoring and control and for transaction processing. | Car airbags, aircraft systems, medical monitors, ticket booking (prevents double booking) |
| Network OS | Resource sharing (files, printers); user management (accounts, permissions); communication between devices. | Windows Server, Linux servers |
| Mobile | For smartphones and tablets; touch interface; lower processing requirements; designed to increase battery life (e.g. closes background apps). | Android, iOS |
Utilities (7.2.2)
| Utility | What it does |
|---|---|
| File management | Create, copy, move, rename, delete and organise files and folders; search; set permissions. |
| Defragmenter | Rearranges fragmented files on a magnetic hard disk so each file's parts are stored together, reducing read head movement and speeding up access. Not needed (and not recommended) for SSDs. |
| File compression | Reduces file size (e.g. ZIP) to save storage and speed up transfers. Lossless keeps all data; lossy (JPEG, MP3) removes some. |
| Package manager | Installs, updates and removes software and its dependencies automatically from a repository, e.g. pip for Python, apt for Linux, winget for Windows. |
| Protection software | Anti-malware and firewalls: scan for and remove threats, block unauthorised connections. |
| Backup software | Copies data on a schedule (full, incremental, differential) so it can be restored after loss. |
Code development tools (7.2.3)
| Tool | Features |
|---|---|
| IDE: code editing tools | Syntax highlighting, auto-complete, auto-indentation, line numbers, error highlighting as you type, refactoring tools. |
| IDE: debugging tools | Breakpoints, stepping through code line by line, watching variable values, call stack view. |
| IDE: screen design tools | Drag-and-drop design of graphical user interfaces (forms, buttons, layouts) that generates the code. |
| Compiler | Translates the whole program into machine code before it runs, producing an executable. Fast to run; code is hidden from users; errors are reported after compiling the whole program; must recompile for each platform. Examples: C, C++, (C# compiles to an intermediate form). |
| Interpreter | Translates and runs code one line at a time. Easier debugging (stops at the error line); runs on any platform with the interpreter; slower to run; source code must be distributed. Example: Python. |
Application software (7.2.4)
| Application | Features and business uses |
|---|---|
| Word processors | Create and format documents: letters, reports, contracts; templates, mail merge, track changes, collaboration. |
| Spreadsheets | Calculations with formulas and functions, charts, sorting and filtering data; budgets, sales analysis, forecasts. |
| Databases | Store large amounts of structured data in tables; queries, forms and reports; customer records, stock control. Reduce duplication compared with spreadsheets. |
| Send messages and attachments; calendars and contacts; formal written record of communication. | |
| Project management software | Plan tasks, Gantt charts, assign resources, track progress and deadlines, Kanban boards (e.g. Microsoft Project, Trello, Jira). |
- Operating system
- Software that manages hardware and resources and provides a platform for applications.
- Time-slicing
- Giving each task a short slot of processor time in turn, so several tasks appear to run at once.
- Interrupt
- A signal to the processor that something needs attention immediately.
- Utility
- System software that maintains or optimises the computer, e.g. backup, compression.
- Package manager
- A tool that installs, updates and removes software and its dependencies, e.g. pip, apt.
- IDE
- Integrated development environment: software that combines a code editor, debugger and other tools for programming.
- Compiler
- Translates the whole source code into machine code in one go, producing an executable file.
- Interpreter
- Translates and runs source code one line at a time.
7.3 Networks
Benefits and drawbacks of networks (7.3.1)
| Benefits | Drawbacks |
|---|---|
|
|
Features of network types (7.3.2)
Networks differ in number of users, connection media (the cables or wireless used) and coverage (the area served).
| Type | Coverage | Example | Notes |
|---|---|---|---|
| PAN: personal area network | A few metres around one person | Phone connected to smartwatch and earbuds by Bluetooth | Very few devices; low power |
| LAN: local area network | One building or site | A college or office network | Owned and managed by the organisation; fast; Ethernet and Wi-Fi |
| MAN: metropolitan area network | A town or city | A council linking its offices and libraries; city-wide Wi-Fi | Often uses fibre leased from telecoms providers |
| WAN: wide area network | Countries or the world | The internet; a bank linking branches nationally | Uses third-party infrastructure; slower and more expensive than LAN |
| VPN: virtual private network | Any distance, over the internet | Staff connecting securely to the office network from home | Creates an encrypted tunnel over a public network so traffic can't be read |
Connectivity methods (7.3.3)
| Method | Features | Benefits | Drawbacks |
|---|---|---|---|
| Copper / Ethernet (twisted pair, e.g. Cat 6) | Electrical signals; up to 100 m per run | Cheap; reliable; fast (1–10 Gbps); easy to install | Interference (EMI); limited distance; signal degrades |
| Fibre-optic | Pulses of light through glass or plastic | Very high bandwidth; long distances; immune to electrical interference; hard to tap | Expensive; fragile; specialist installation |
| Wireless access points (Wi-Fi) | Radio waves; WAP connects wireless devices to the wired network | Mobility; no cables; easy to add devices | Slower and less reliable than cable; range limited by walls; interference; must be secured (WPA3) |
Topologies (7.3.4)
| Topology | Layout | Benefits | Drawbacks |
|---|---|---|---|
| Star | Every device connects to a central switch. | One cable failure only affects one device; easy to add devices; good performance | If the central switch fails, the whole network fails; lots of cable |
| Mesh | Devices connect to many others (full mesh: every device to every other). | Very resilient: many routes for data; no single point of failure | Expensive and complex; lots of connections. Partial mesh reduces this. Used in WANs and the internet backbone, and wireless mesh home Wi-Fi. |
| Tree | Several star networks connected in a hierarchy from a root. | Scalable; easy to manage in sections (e.g. one star per floor) | If the root or a backbone link fails, the sections below are cut off |
Physical topology is how devices are actually cabled. Logical topology is how data actually flows. For example, a network can be physically wired as a star but data may be passed in a logical bus or ring pattern; Wi-Fi devices have no cables but logically form a star around the access point.
Network models (7.3.5)
| Model | How it works | Benefits | Drawbacks |
|---|---|---|---|
| Client-server | Clients request services from central servers that store data and manage security. | Central backup and security; easy to manage users; scalable | Servers are expensive; needs IT staff; server failure affects everyone |
| Thin client | Clients have minimal processing power and storage; the server does almost all processing (e.g. virtual desktops). | Cheap clients; easy to manage centrally; data never stored on the client so more secure | Totally dependent on the server and network; poor for heavy graphics |
| Peer-to-peer | All computers are equal and share resources directly; no central server. | Cheap; easy to set up; no single point of failure | No central backup or security; harder to manage as it grows; performance depends on each peer |
Network components (7.3.6)
| Component | Role |
|---|---|
| Server | Provides services to clients: files, web pages, email, authentication, printing. |
| Client | A device that requests and uses services from a server. |
| Router | Connects different networks (e.g. a LAN to the internet) and forwards packets towards their destination using IP addresses. Home routers often include a switch, WAP and firewall. |
| Network switch | Connects devices within a LAN; learns MAC addresses and sends data only to the intended device, reducing unnecessary traffic. |
| Internet connection / backbone | The connection to an ISP; the backbone is the high-capacity fibre network of main routes that carries internet traffic between countries and providers. |
- Topology
- The arrangement of devices and connections in a network.
- Router
- Forwards data packets between different networks using IP addresses.
- Switch
- Connects devices within a LAN and forwards data only to the device it is addressed to, using MAC addresses.
7.3.7–8 Network models: OSI and TCP/IP
Layered models split networking into separate layers. Each layer has a job and passes data to the layer above or below. Benefits: hardware and software from different vendors work together; one layer can change without affecting the others; easier to troubleshoot.
| OSI layer | Function | Example protocols / devices | TCP/IP layer |
|---|---|---|---|
| 7 Application | Provides network services to applications the user interacts with. | HTTP, HTTPS, FTP, SMTP, POP, IMAP, DNS, DHCP | Application |
| 6 Presentation | Formats, encrypts and compresses data so it can be understood by the receiver. | TLS/SSL encryption, JPEG, ASCII, MP4 | Application |
| 5 Session | Opens, manages and closes communication sessions between devices. | NetBIOS, RPC, session handling in TLS | Application |
| 4 Transport | Splits data into segments, end-to-end delivery, error checking, flow control, port numbers. | TCP (reliable, ordered), UDP (fast, no guarantee) | Transport |
| 3 Network | Logical (IP) addressing and routing packets between networks. | IP, ICMP, RIP, OSPF; routers | Internet |
| 2 Data link | Physical (MAC) addressing, framing, error detection on the local link. | Ethernet, Wi-Fi (802.11) frames, ARP; switches | Network (access) |
| 1 Physical | Transmits raw bits as electrical, light or radio signals. | Cables, connectors, radio frequencies; hubs, NICs | Network (access) |
★ REMEMBERING THE OSI LAYERS
From layer 7 down: All People Seem To Need Data Processing. From layer 1 up: Please Do Not Throw Sausage Pizza Away.
| TCP/IP layer | Function | Protocols |
|---|---|---|
| Application | Combines OSI 5–7. Provides services to applications and formats data. | HTTP, HTTPS, FTP, SFTP, SMTP, POP, IMAP, DNS, DHCP |
| Transport | Establishes end-to-end connections, splits data into numbered segments, uses ports, handles retransmission. | TCP, UDP |
| Internet | Adds source and destination IP addresses and routes packets across networks. | IP (IPv4, IPv6), ICMP |
| Network (access / link) | Combines OSI 1–2. Sends frames over the physical medium using MAC addresses. | Ethernet, Wi-Fi, ARP |
Practise: which OSI layer?
A router chooses the next hop for a packet using its IP address.
Data is encrypted with TLS and compressed before sending.
A switch reads a MAC address and forwards the frame to one port.
TCP numbers segments and resends one that went missing.
Bits travel as pulses of light down a fibre cable.
A browser sends an HTTPS request for a web page.
A connection between two devices is opened, kept alive and closed.
- Protocol
- A set of rules that govern how data is transmitted and received.
7.3.9–11 Packets, protocols, bandwidth and latency
Data packets (7.3.9)
| Part | Contents | Role |
|---|---|---|
| Header | Source IP address; destination IP address; packet sequence number; total number of packets; protocol; time to live (TTL) | Gets the packet to the right place and lets the receiver reassemble packets in order. TTL stops lost packets circulating forever. |
| Payload | The actual data being sent (a chunk of the file, email or web page) | The content. |
| Trailer | Error-checking value (e.g. CRC) and end-of-packet marker | Lets the receiver detect whether the packet was corrupted. |
Packet switching
- Data is split into packets, each with a header.
- Each packet is sent independently; routers forward each one along the best available route at that moment.
- Packets may take different routes and arrive out of order.
- The receiver uses the sequence numbers to reassemble them.
- Missing or corrupted packets are requested again (TCP).
Benefits: efficient use of the network; routes around failures and congestion. Drawbacks: packets can arrive out of order or be delayed, which matters for real-time traffic like video calls.
Causes of packet loss
- Network congestion: routers' buffers fill up and packets are dropped.
- Faulty or damaged hardware (cables, routers, NICs).
- Wireless interference or weak signal.
- TTL reaching zero before arrival.
- Software bugs or misconfigured devices; deliberate attacks such as DoS.
Error handling: cyclic redundancy check (CRC)
- The sender performs a calculation (a form of polynomial division) on the packet's data to produce a CRC value.
- The CRC value is added to the trailer.
- The receiver performs the same calculation on the data it received.
- If the values match, the data is accepted. If not, the packet is corrupt and is discarded and re-requested.
Network protocols (7.3.10)
| Protocol | Purpose | Port (useful to know) |
|---|---|---|
| HTTP | HyperText Transfer Protocol: requests and transfers web pages between browser and web server. Not encrypted. | 80 |
| HTTPS | HTTP Secure: HTTP encrypted with TLS, so data such as passwords and card numbers cannot be read if intercepted; server identity verified with a certificate. | 443 |
| SMTP | Simple Mail Transfer Protocol: sends email from client to server and between mail servers. | 25 / 587 |
| POP (POP3) | Post Office Protocol: downloads email from the server to one device, usually deleting it from the server. Good for one device and offline access. | 110 |
| IMAP | Internet Message Access Protocol: email stays on the server and is synchronised across all devices. | 143 |
| RIP | Routing Information Protocol: routers share routes and choose the path with the fewest hops (max 15). Simple; suited to small networks; slow to adapt. | - |
| OSPF | Open Shortest Path First: routers share link-state information and calculate the fastest path based on cost (e.g. bandwidth). Scales to large networks; adapts quickly to changes. | - |
| FTP | File Transfer Protocol: uploads and downloads files between client and server. Not encrypted. | 20 / 21 |
| SFTP | SSH File Transfer Protocol: secure, encrypted file transfer. | 22 |
| DHCP | Dynamic Host Configuration Protocol: automatically gives devices an IP address, subnet mask, default gateway and DNS server when they join a network. | 67 / 68 |
| DNS | Domain Name System: translates domain names (wakefield.ac.uk) into IP addresses. | 53 |
Practise: which protocol?
Your laptop joins the college Wi-Fi and is automatically given an IP address.
You read an email on your phone and it shows as read on your laptop too.
Your browser finds the IP address for neocities.org.
Your email client sends a message to the mail server.
A developer uploads website files over an encrypted connection.
A customer enters card details on a shop's checkout page.
Routers in a large network calculate the fastest path using link cost.
Routers in a small network pick the route with the fewest hops.
Email is downloaded to one computer and removed from the server.
Bandwidth and latency (7.3.11)
| Bandwidth | Latency | |
|---|---|---|
| Meaning | The maximum amount of data that can be transferred per second | The delay for data to travel from sender to receiver |
| Unit | Bits per second (Mbps, Gbps) | Milliseconds (ms) |
| Affected by | Connection type, number of users sharing it, ISP package | Distance, number of hops/routers, congestion, connection type (satellite is high) |
| Effect when poor | Slow downloads and uploads, video buffering, low quality streams | Lag in online games and video calls, slow response to clicks, even if downloads are fast |
✎ WORKED EXAMPLE: DOWNLOAD TIME
- How long will a 500 MB file take to download on a 100 Mbps connection?
- Convert megabytes to megabits: 500 × 8 = 4,000 Mb.
- Time = data ÷ bandwidth = 4,000 ÷ 100 = 40 seconds (in ideal conditions; real throughput is usually lower).
Try it: download time calculator
- Packet
- A small unit of data sent across a network, containing a header, payload and trailer.
- Packet switching
- Splitting data into packets that travel independently across a network and are reassembled at the destination.
- CRC
- Cyclic redundancy check: an error-detection value calculated from the packet's data and checked by the receiver.
- Bandwidth
- The maximum amount of data that can be transferred per second, measured in bits per second.
- Latency
- The time delay for data to travel from source to destination, measured in milliseconds.
7.4 Virtual environments
Components (7.4.1)
| Component | Role and characteristics |
|---|---|
| Virtual machine: clients | Virtual PCs (a full desktop OS running as software), virtual switches (connect VMs to each other and to the physical network), virtual routers (route traffic between virtual networks). |
| Virtual machine: servers | Several virtual servers (web, database, email) run on one physical server, each with its own OS, isolated from the others. |
| Hypervisor: type 1 (bare metal) | Runs directly on the hardware with no host OS. Efficient, secure, high performance. Used in data centres. Examples: VMware ESXi, Microsoft Hyper-V, Proxmox. |
| Hypervisor: type 2 (hosted) | Runs as an application on a host OS. Easy to install; slower because it goes through the host OS. Used for testing and learning. Examples: VirtualBox, VMware Workstation. |
↓
Hypervisor
↓
Hardware
↓
Hypervisor (an app)
↓
Host OS
↓
Hardware
Key features (7.4.2)
| Feature | Meaning |
|---|---|
| Increased security | VMs are separated, so malware in one VM does not easily spread to others or the host; snapshots allow quick recovery. |
| Managed execution | The hypervisor controls each VM's access to resources (CPU, RAM, storage), so VMs run in a controlled way. |
| Sharing | Several VMs share one set of physical hardware. |
| Aggregation | Resources from several physical machines are combined to act as one larger pool (e.g. a cluster). |
| Emulation | A VM can imitate different hardware or an older OS, so legacy software can still run. |
| Isolation | Each VM is independent: a crash in one does not affect the others. |
| Portability | A VM is stored as files, so it can be copied, backed up or moved to another host easily. |
Benefits and drawbacks (7.4.3–7.4.4)
| Benefits | Drawbacks |
|---|---|
|
|
- Virtual machine (VM)
- A software-based computer that runs an operating system and applications like a physical machine.
- Hypervisor
- Software that creates and runs virtual machines and shares physical hardware between them.
7.5 Cloud environments
Types of cloud (7.5.1)
| Private cloud | Public cloud | |
|---|---|---|
| Meaning | Cloud infrastructure used by one organisation only, on its own premises or hosted by a provider. | Infrastructure owned by a provider and shared by many customers over the internet. |
| Examples | A bank's own data centre running virtualised services | AWS, Microsoft Azure, Google Cloud |
| Benefits | Full control; greater security and privacy; can meet strict regulations | No upfront hardware cost; pay as you go; scales easily; provider handles maintenance |
| Drawbacks | Expensive to build and maintain; needs skilled staff; limited scalability | Less control; data held by a third party (possibly abroad); depends on internet connection; costs can grow |
Benefits of cloud (7.5.2)
| Benefit | Meaning |
|---|---|
| Portability | Access services and data from any device, anywhere with internet; move workloads between environments. |
| Elasticity | Resources scale up automatically at busy times and down when quiet, e.g. an online shop on Black Friday. |
| Fewer storage limitations | Storage can be increased on demand without buying hardware. |
| Cost effectiveness | No upfront hardware purchase; pay only for what is used; fewer IT staff needed for maintenance. |
Delivery models (7.5.3)
| Layer | On-premises | IaaS | PaaS | SaaS |
|---|---|---|---|---|
| User accounts | Client | Client | Client | Client |
| Data | Client | Client | Client | Client |
| Application software | Client | Client | Client | Provider |
| Runtime | Client | Client | Provider | Provider |
| System software (middleware and OS) | Client | Client | Provider | Provider |
| Virtualisation | Client | Provider | Provider | Provider |
| Hardware (servers, network, storage) | Client | Provider | Provider | Provider |
| Model | Example | Advantages | Disadvantages | Suits |
|---|---|---|---|---|
| IaaS | AWS EC2, Azure Virtual Machines | Most control and flexibility; no hardware to buy | Client must manage and secure OS and software; needs skilled staff | IT teams who need custom servers |
| PaaS | Azure App Service, Google App Engine, Heroku | Developers focus on code; no OS patching; quick deployment | Less control; limited to supported languages/tools; vendor lock-in | Software developers building and deploying apps |
| SaaS | Microsoft 365, Google Workspace, Salesforce | Ready to use; no installation; provider handles everything | Least control and customisation; data held by provider; subscription costs continue | End users and businesses needing standard software |
★ EXAM TIP
Learn the responsibility table. A common question gives a scenario and asks who is responsible for, say, patching the operating system. Answer: client in IaaS, provider in PaaS and SaaS.
Practise: who is responsible?
Patching the operating system on an IaaS virtual server.
Patching the operating system under a PaaS web app.
Managing user accounts in a SaaS email service.
Replacing a failed hard drive in an IaaS data centre.
Writing and updating the application code on PaaS.
Updating the application software in SaaS.
- Elasticity
- The ability of cloud resources to scale up or down automatically to match demand.
- IaaS
- Infrastructure as a Service: the provider supplies virtualised hardware; the client manages everything from the OS up.
- PaaS
- Platform as a Service: the provider supplies hardware, OS and runtime; the client manages applications and data.
- SaaS
- Software as a Service: the provider supplies the complete application; the client manages users and data.
7.6 Resilient digital environments
Benefits of resilience (7.6.1)
| Benefit | Impact on organisation and clients |
|---|---|
| Increased security: data security in storage and transfer; reduced vulnerabilities | Data is protected from loss and attack; fewer weaknesses for attackers to exploit. |
| Increased reputation: protect brand/image; retain customer confidence | Customers trust an organisation whose services are reliable and whose data is safe. |
| Reduction in downtime | Services stay available, so sales, productivity and customer service continue. |
Methods to improve resilience (7.6.2)
| Method | Detail |
|---|---|
| Software updates / upgrades | Planned updates and upgrades on a schedule (tested first, outside busy hours). Patches released quickly in response to new vulnerabilities. |
| Hardware replacement | Rolling replacement plans: replace a proportion of hardware each year so nothing becomes too old and costs are spread. Secure disposal: wipe or physically destroy drives so data cannot be recovered; use certified recyclers. |
| Data and system redundancy | Duplicate components so one failure doesn't stop the service: RAID disks, spare power supplies, UPS, backup internet connection, duplicate servers with failover, data replicated to another site. |
| Device hardening | Remove unneeded ports, applications, permissions and access; change default passwords; disable unused services. Fewer ways in for attackers. |
| Backup systems and recovery procedures | Onsite: fast to restore but at risk from the same fire or theft. Remote/offsite: protected from local disasters but slower to restore. Cloud: automatic, offsite, scalable; depends on internet and provider. Recovery procedures must be documented and tested. |
| Hot, warm and cold sites | Alternative locations to continue operating after a disaster (table below). |
| Standard operating procedures | Written step-by-step procedures so tasks are done consistently and safely: effective staff training; induction for new staff; procedures for new digital systems; procedures for new or updated policies. |
| Site | What it has | Time to recover | Cost |
|---|---|---|---|
| Hot | Fully equipped, running systems with real-time or near real-time copies of data | Minutes to hours | Highest |
| Warm | Hardware and network connections in place; data must be restored from backup | Hours to days | Medium |
| Cold | Space, power and cooling only; equipment must be brought in and set up | Days to weeks | Lowest |
✎ WORKED EXAMPLE
An online bank cannot afford any downtime, so a hot site is justified despite the cost. A small accountancy firm could use cloud backups and a cold site (or simply work from home), because a few days' disruption is tolerable and the cost of a hot site would be out of proportion.
- Redundancy
- Having duplicate components or data so that if one fails, another takes over.
- Device hardening
- Reducing a device's attack surface by removing unneeded ports, applications, permissions and access.
- Hot site
- A fully equipped backup site with up-to-date data that can take over almost immediately.
- Cold site
- A backup location with power and space but no equipment or data set up.
- Standard operating procedure (SOP)
- A written, step-by-step instruction for carrying out a routine task consistently.
Knowledge organiser
COMPUTERS AND HARDWARE
- PC, mobile, server, embedded
- CPU: cores, clock speed, cache, mobile chips
- RAM volatile; ROM non-volatile
- Storage: magnetic, solid state, optical
- GPU, network interface, cooling, sensors
OPERATING SYSTEMS
- Batch: high volume, scheduled
- Multitasking: time-slicing, interrupts
- Real-time: guaranteed response
- Network: sharing, users
- Mobile: battery life
SOFTWARE TOOLS
- Utilities: file management, defragmenter, compression, package manager, protection, backup
- IDE: editor, debugger, screen designer
- Compiler: whole program; interpreter: line by line
NETWORKS
- PAN, LAN, MAN, WAN, VPN
- Copper, fibre, wireless
- Star, mesh, tree; logical vs physical
- Client-server, thin client, peer-to-peer
- Server, client, router, switch, backbone
LAYER MODELS
- OSI: application, presentation, session, transport, network, data link, physical
- TCP/IP: application, transport, internet, network
- Packet: header, payload, trailer; CRC detects errors
PROTOCOLS
- HTTP 80 / HTTPS 443
- SMTP send; POP download; IMAP sync
- RIP hops; OSPF fastest path
- FTP / SFTP, DHCP, DNS
- Bandwidth = capacity; latency = delay
VIRTUAL AND CLOUD
- Type 1 hypervisor (bare metal), type 2 (hosted)
- Features: security, managed execution, sharing, aggregation, emulation, isolation, portability
- Private vs public cloud; elasticity
- IaaS / PaaS / SaaS: client manages less as you go along
RESILIENCE
- Planned updates and patches
- Rolling hardware replacement; secure disposal
- Redundancy; device hardening
- Backups onsite / offsite / cloud
- Hot, warm, cold sites; standard operating procedures
Flashcards
Every Quick recall definition on this page. Click the card to flip it; use the arrow keys to move.
Quiz me
Random questions on CA7 from the course question bank. Pick a topic or question type, answer in your head or on paper, then check. Your score is saved on this device.
Exam-style questions
Allow about 100 minutes.
State two features of an embedded device.
- Any two: performs one dedicated function; built into a larger device; low power; small; firmware stored in ROM; limited or no user interface; often real-time.
Explain why increasing the cache size of a processor can improve performance.
- Cache stores frequently used instructions/data close to the CPU (1) so the CPU waits less for data from slower RAM / fewer RAM accesses (1).
Compare RAM and ROM.
- RAM is volatile; ROM is non-volatile (1).
- RAM can be read and written; ROM is normally read only (1).
- RAM holds programs and data currently in use; ROM holds startup instructions / firmware / BIOS (1).
- RAM is usually much larger than ROM (1).
A delivery company needs laptops for drivers who work outdoors. Explain why solid state storage is more suitable than magnetic storage.
- Point (1) + linked explanation (1), twice. E.g. no moving parts (1) so less likely to be damaged if the laptop is dropped or vibrates in a van (1); uses less power (1) so battery lasts longer on shift (1); faster access (1) so apps load quickly at each delivery (1).
Describe how a multitasking operating system allows several programs to appear to run at the same time.
- The CPU switches rapidly between tasks (1)
- each task is given a time slice (1)
- switching is so fast that programs appear to run simultaneously; interrupts allow urgent tasks to take priority (1).
Explain two differences between a compiler and an interpreter.
- Difference (1) + expansion (1), twice. E.g. a compiler translates the whole program at once (1) producing an executable, whereas an interpreter translates and runs line by line (1); compiled code runs faster (1) because translation is done in advance (1); interpreter stops at the first error (1) making debugging easier (1); compiled program can be distributed without source code (1) whereas interpreted needs the source and interpreter (1).
State the purpose of each protocol: (a) DHCP (b) DNS (c) IMAP.
- (a) Automatically assigns IP addresses (and other settings) to devices joining a network (1).
- (b) Translates domain names into IP addresses (1).
- (c) Retrieves email while keeping it on the server, synchronised across devices (1).
Describe the role of the transport layer in the TCP/IP model.
- Any three: splits data into segments/packets; numbers them so they can be reassembled; establishes the connection between hosts (TCP); uses port numbers to direct data to the correct application; checks for errors and requests retransmission; flow control.
Explain how a cyclic redundancy check is used to detect errors in a data packet.
- Sender calculates a CRC value from the data and adds it to the packet trailer (1).
- Receiver performs the same calculation on the received data (1).
- If the values do not match the packet is corrupt and is discarded / re-sent (1).
A user has a fast broadband connection but experiences lag in online games. Explain the likely cause.
- High latency (1) caused by e.g. distance to the game server, congestion, Wi-Fi interference or many hops, so there is a delay even though bandwidth is high (1).
Calculate how long it would take to transfer a 2 GB file over a 50 Mbps connection. Show your working. (1 GB = 1,000 MB)
- 2 GB = 2,000 MB = 16,000 Mb (1)
- 16,000 ÷ 50 (1)
- = 320 seconds (5 minutes 20 seconds) (1). Accept 2 GB = 2,048 MB giving 327.68 seconds.
Explain the difference between a type 1 and a type 2 hypervisor.
- Type 1 runs directly on the hardware with no host OS (1); type 2 runs as an application on top of a host OS (1).
- Type 1 is more efficient/faster/more secure, used in data centres (1); type 2 is easier to set up, used for testing/learning on a PC (1).
A start-up wants to deploy a web application without managing servers or operating systems. Identify the most suitable cloud delivery model and justify your choice.
- PaaS (1).
- The provider manages hardware, virtualisation, OS and runtime (1)
- so the start-up only manages its application code and data, saving time and needing fewer skilled staff (1).
Explain two methods a hospital could use to improve the resilience of its patient record system.
- Method (1) + linked explanation (1), twice. E.g. redundancy: duplicate servers with failover (1) so records are still available if one fails (1); regular backups stored offsite/cloud (1) so data can be restored after ransomware or fire (1); hot site (1) so service continues within minutes after a disaster (1); patching (1) to close vulnerabilities (1); device hardening (1); SOPs and staff training (1).
A college with three buildings is redesigning its network. Discuss the suitability of star and mesh topologies for the network.
Indicative content:
- Star within each building: easy to add devices; one cable fault affects only one device; good performance; but the central switch is a single point of failure.
- Mesh between buildings: multiple routes, so if one link between buildings fails traffic is rerouted; high resilience; but more expensive cabling (fibre between buildings) and more complex.
- A combination is likely: star in each building, partial mesh or tree linking them. Tree is simpler and cheaper but the root is a single point of failure.
- Consider budget, need for resilience (e.g. exams, online learning), and scalability.
A national chain of 40 gyms runs its booking, membership and payment systems on old servers in its head office. It is considering moving to a public cloud using SaaS for membership and IaaS for its booking system. Evaluate this plan.
◆ SUPPORT: SENTENCE STARTERS
- A star topology would suit each building because ... However, ...
- A mesh would provide more routes between buildings, which means ...
- Moving to the cloud would give the gym chain elasticity, so at busy times such as January ...
- With IaaS the gym chain would still be responsible for ...
- A risk of the public cloud is ...
Indicative content:
- Benefits: no need to replace old servers; elasticity for busy periods (January, evening bookings); access from all 40 gyms and members' phones; provider's data centres offer redundancy and backups, improving resilience; predictable subscription costs; SaaS membership software is maintained and updated by the provider.
- Shared responsibility: with IaaS the gym chain still manages the OS, runtime, booking application, data and user accounts, so it needs skilled staff; with SaaS it only manages users and data.
- Drawbacks: dependent on internet connection at every gym; ongoing costs may exceed owning servers over time; data held by a third party, possibly abroad, so data protection checks needed (payment data must meet PCI DSS); migration risk and staff training; vendor lock-in; less control over SaaS customisation.
- Could suggest PaaS for the booking system to reduce management burden.
- Judgement: e.g. the plan is suitable because old servers are a resilience risk and cloud offers scalability and redundancy, provided the chain has reliable internet at each site and checks where data is stored.
▲ STRETCH
- For question 16, include the shared responsibility model and explain how the plan affects resilience (redundancy, backups, downtime).
- Explain why OSPF is more suitable than RIP for a large organisation's network.